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    COPY = 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_CONDITION = auto()
 296    MATCH_RECOGNIZE = auto()
 297    MEMBER_OF = auto()
 298    MERGE = auto()
 299    MOD = auto()
 300    MODEL = auto()
 301    NATURAL = auto()
 302    NEXT = auto()
 303    NOTNULL = auto()
 304    NULL = auto()
 305    OBJECT_IDENTIFIER = auto()
 306    OFFSET = auto()
 307    ON = auto()
 308    ONLY = auto()
 309    OPERATOR = auto()
 310    ORDER_BY = auto()
 311    ORDER_SIBLINGS_BY = auto()
 312    ORDERED = auto()
 313    ORDINALITY = auto()
 314    OUTER = auto()
 315    OVER = auto()
 316    OVERLAPS = auto()
 317    OVERWRITE = auto()
 318    PARTITION = auto()
 319    PARTITION_BY = auto()
 320    PERCENT = auto()
 321    PIVOT = auto()
 322    PLACEHOLDER = auto()
 323    POSITIONAL = auto()
 324    PRAGMA = auto()
 325    PREWHERE = auto()
 326    PRIMARY_KEY = auto()
 327    PROCEDURE = auto()
 328    PROPERTIES = auto()
 329    PSEUDO_TYPE = auto()
 330    QUALIFY = auto()
 331    QUOTE = auto()
 332    RANGE = auto()
 333    RECURSIVE = auto()
 334    REFRESH = auto()
 335    REPLACE = auto()
 336    RETURNING = auto()
 337    REFERENCES = auto()
 338    RIGHT = auto()
 339    RLIKE = auto()
 340    ROLLBACK = auto()
 341    ROLLUP = auto()
 342    ROW = auto()
 343    ROWS = auto()
 344    SELECT = auto()
 345    SEMI = auto()
 346    SEPARATOR = auto()
 347    SEQUENCE = auto()
 348    SERDE_PROPERTIES = auto()
 349    SET = auto()
 350    SETTINGS = auto()
 351    SHOW = auto()
 352    SIMILAR_TO = auto()
 353    SOME = auto()
 354    SORT_BY = auto()
 355    START_WITH = auto()
 356    STORAGE_INTEGRATION = auto()
 357    STRUCT = auto()
 358    TABLE_SAMPLE = auto()
 359    TEMPORARY = auto()
 360    TOP = auto()
 361    THEN = auto()
 362    TRUE = auto()
 363    TRUNCATE = auto()
 364    UNCACHE = auto()
 365    UNION = auto()
 366    UNNEST = auto()
 367    UNPIVOT = auto()
 368    UPDATE = auto()
 369    USE = auto()
 370    USING = auto()
 371    VALUES = auto()
 372    VIEW = auto()
 373    VOLATILE = auto()
 374    WHEN = auto()
 375    WHERE = auto()
 376    WINDOW = auto()
 377    WITH = auto()
 378    UNIQUE = auto()
 379    VERSION_SNAPSHOT = auto()
 380    TIMESTAMP_SNAPSHOT = auto()
 381    OPTION = auto()
 382
 383
 384_ALL_TOKEN_TYPES = list(TokenType)
 385_TOKEN_TYPE_TO_INDEX = {token_type: i for i, token_type in enumerate(_ALL_TOKEN_TYPES)}
 386
 387
 388class Token:
 389    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
 390
 391    @classmethod
 392    def number(cls, number: int) -> Token:
 393        """Returns a NUMBER token with `number` as its text."""
 394        return cls(TokenType.NUMBER, str(number))
 395
 396    @classmethod
 397    def string(cls, string: str) -> Token:
 398        """Returns a STRING token with `string` as its text."""
 399        return cls(TokenType.STRING, string)
 400
 401    @classmethod
 402    def identifier(cls, identifier: str) -> Token:
 403        """Returns an IDENTIFIER token with `identifier` as its text."""
 404        return cls(TokenType.IDENTIFIER, identifier)
 405
 406    @classmethod
 407    def var(cls, var: str) -> Token:
 408        """Returns an VAR token with `var` as its text."""
 409        return cls(TokenType.VAR, var)
 410
 411    def __init__(
 412        self,
 413        token_type: TokenType,
 414        text: str,
 415        line: int = 1,
 416        col: int = 1,
 417        start: int = 0,
 418        end: int = 0,
 419        comments: t.Optional[t.List[str]] = None,
 420    ) -> None:
 421        """Token initializer.
 422
 423        Args:
 424            token_type: The TokenType Enum.
 425            text: The text of the token.
 426            line: The line that the token ends on.
 427            col: The column that the token ends on.
 428            start: The start index of the token.
 429            end: The ending index of the token.
 430            comments: The comments to attach to the token.
 431        """
 432        self.token_type = token_type
 433        self.text = text
 434        self.line = line
 435        self.col = col
 436        self.start = start
 437        self.end = end
 438        self.comments = [] if comments is None else comments
 439
 440    def __repr__(self) -> str:
 441        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
 442        return f"<Token {attributes}>"
 443
 444
 445class _Tokenizer(type):
 446    def __new__(cls, clsname, bases, attrs):
 447        klass = super().__new__(cls, clsname, bases, attrs)
 448
 449        def _convert_quotes(arr: t.List[str | t.Tuple[str, str]]) -> t.Dict[str, str]:
 450            return dict(
 451                (item, item) if isinstance(item, str) else (item[0], item[1]) for item in arr
 452            )
 453
 454        def _quotes_to_format(
 455            token_type: TokenType, arr: t.List[str | t.Tuple[str, str]]
 456        ) -> t.Dict[str, t.Tuple[str, TokenType]]:
 457            return {k: (v, token_type) for k, v in _convert_quotes(arr).items()}
 458
 459        klass._QUOTES = _convert_quotes(klass.QUOTES)
 460        klass._IDENTIFIERS = _convert_quotes(klass.IDENTIFIERS)
 461
 462        klass._FORMAT_STRINGS = {
 463            **{
 464                p + s: (e, TokenType.NATIONAL_STRING)
 465                for s, e in klass._QUOTES.items()
 466                for p in ("n", "N")
 467            },
 468            **_quotes_to_format(TokenType.BIT_STRING, klass.BIT_STRINGS),
 469            **_quotes_to_format(TokenType.BYTE_STRING, klass.BYTE_STRINGS),
 470            **_quotes_to_format(TokenType.HEX_STRING, klass.HEX_STRINGS),
 471            **_quotes_to_format(TokenType.RAW_STRING, klass.RAW_STRINGS),
 472            **_quotes_to_format(TokenType.HEREDOC_STRING, klass.HEREDOC_STRINGS),
 473            **_quotes_to_format(TokenType.UNICODE_STRING, klass.UNICODE_STRINGS),
 474        }
 475
 476        klass._STRING_ESCAPES = set(klass.STRING_ESCAPES)
 477        klass._IDENTIFIER_ESCAPES = set(klass.IDENTIFIER_ESCAPES)
 478        klass._COMMENTS = {
 479            **dict(
 480                (comment, None) if isinstance(comment, str) else (comment[0], comment[1])
 481                for comment in klass.COMMENTS
 482            ),
 483            "{#": "#}",  # Ensure Jinja comments are tokenized correctly in all dialects
 484        }
 485
 486        klass._KEYWORD_TRIE = new_trie(
 487            key.upper()
 488            for key in (
 489                *klass.KEYWORDS,
 490                *klass._COMMENTS,
 491                *klass._QUOTES,
 492                *klass._FORMAT_STRINGS,
 493            )
 494            if " " in key or any(single in key for single in klass.SINGLE_TOKENS)
 495        )
 496
 497        if USE_RS_TOKENIZER:
 498            settings = RsTokenizerSettings(
 499                white_space={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.WHITE_SPACE.items()},
 500                single_tokens={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.SINGLE_TOKENS.items()},
 501                keywords={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.KEYWORDS.items()},
 502                numeric_literals=klass.NUMERIC_LITERALS,
 503                identifiers=klass._IDENTIFIERS,
 504                identifier_escapes=klass._IDENTIFIER_ESCAPES,
 505                string_escapes=klass._STRING_ESCAPES,
 506                quotes=klass._QUOTES,
 507                format_strings={
 508                    k: (v1, _TOKEN_TYPE_TO_INDEX[v2])
 509                    for k, (v1, v2) in klass._FORMAT_STRINGS.items()
 510                },
 511                has_bit_strings=bool(klass.BIT_STRINGS),
 512                has_hex_strings=bool(klass.HEX_STRINGS),
 513                comments=klass._COMMENTS,
 514                var_single_tokens=klass.VAR_SINGLE_TOKENS,
 515                commands={_TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMANDS},
 516                command_prefix_tokens={
 517                    _TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMAND_PREFIX_TOKENS
 518                },
 519                heredoc_tag_is_identifier=klass.HEREDOC_TAG_IS_IDENTIFIER,
 520            )
 521            token_types = RsTokenTypeSettings(
 522                bit_string=_TOKEN_TYPE_TO_INDEX[TokenType.BIT_STRING],
 523                break_=_TOKEN_TYPE_TO_INDEX[TokenType.BREAK],
 524                dcolon=_TOKEN_TYPE_TO_INDEX[TokenType.DCOLON],
 525                heredoc_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEREDOC_STRING],
 526                raw_string=_TOKEN_TYPE_TO_INDEX[TokenType.RAW_STRING],
 527                hex_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEX_STRING],
 528                identifier=_TOKEN_TYPE_TO_INDEX[TokenType.IDENTIFIER],
 529                number=_TOKEN_TYPE_TO_INDEX[TokenType.NUMBER],
 530                parameter=_TOKEN_TYPE_TO_INDEX[TokenType.PARAMETER],
 531                semicolon=_TOKEN_TYPE_TO_INDEX[TokenType.SEMICOLON],
 532                string=_TOKEN_TYPE_TO_INDEX[TokenType.STRING],
 533                var=_TOKEN_TYPE_TO_INDEX[TokenType.VAR],
 534                heredoc_string_alternative=_TOKEN_TYPE_TO_INDEX[klass.HEREDOC_STRING_ALTERNATIVE],
 535            )
 536            klass._RS_TOKENIZER = RsTokenizer(settings, token_types)
 537        else:
 538            klass._RS_TOKENIZER = None
 539
 540        return klass
 541
 542
 543class Tokenizer(metaclass=_Tokenizer):
 544    SINGLE_TOKENS = {
 545        "(": TokenType.L_PAREN,
 546        ")": TokenType.R_PAREN,
 547        "[": TokenType.L_BRACKET,
 548        "]": TokenType.R_BRACKET,
 549        "{": TokenType.L_BRACE,
 550        "}": TokenType.R_BRACE,
 551        "&": TokenType.AMP,
 552        "^": TokenType.CARET,
 553        ":": TokenType.COLON,
 554        ",": TokenType.COMMA,
 555        ".": TokenType.DOT,
 556        "-": TokenType.DASH,
 557        "=": TokenType.EQ,
 558        ">": TokenType.GT,
 559        "<": TokenType.LT,
 560        "%": TokenType.MOD,
 561        "!": TokenType.NOT,
 562        "|": TokenType.PIPE,
 563        "+": TokenType.PLUS,
 564        ";": TokenType.SEMICOLON,
 565        "/": TokenType.SLASH,
 566        "\\": TokenType.BACKSLASH,
 567        "*": TokenType.STAR,
 568        "~": TokenType.TILDA,
 569        "?": TokenType.PLACEHOLDER,
 570        "@": TokenType.PARAMETER,
 571        "#": TokenType.HASH,
 572        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 573        "'": TokenType.UNKNOWN,
 574        "`": TokenType.UNKNOWN,
 575        '"': TokenType.UNKNOWN,
 576    }
 577
 578    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 579    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 580    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 581    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 582    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 583    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 584    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 585    IDENTIFIER_ESCAPES = ['"']
 586    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 587    STRING_ESCAPES = ["'"]
 588    VAR_SINGLE_TOKENS: t.Set[str] = set()
 589
 590    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 591    HEREDOC_TAG_IS_IDENTIFIER = False
 592
 593    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 594    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 595
 596    # Autofilled
 597    _COMMENTS: t.Dict[str, str] = {}
 598    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 599    _IDENTIFIERS: t.Dict[str, str] = {}
 600    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 601    _QUOTES: t.Dict[str, str] = {}
 602    _STRING_ESCAPES: t.Set[str] = set()
 603    _KEYWORD_TRIE: t.Dict = {}
 604    _RS_TOKENIZER: t.Optional[t.Any] = None
 605
 606    KEYWORDS: t.Dict[str, TokenType] = {
 607        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 608        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 609        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 610        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 611        "/*+": TokenType.HINT,
 612        "==": TokenType.EQ,
 613        "::": TokenType.DCOLON,
 614        "||": TokenType.DPIPE,
 615        ">=": TokenType.GTE,
 616        "<=": TokenType.LTE,
 617        "<>": TokenType.NEQ,
 618        "!=": TokenType.NEQ,
 619        ":=": TokenType.COLON_EQ,
 620        "<=>": TokenType.NULLSAFE_EQ,
 621        "->": TokenType.ARROW,
 622        "->>": TokenType.DARROW,
 623        "=>": TokenType.FARROW,
 624        "#>": TokenType.HASH_ARROW,
 625        "#>>": TokenType.DHASH_ARROW,
 626        "<->": TokenType.LR_ARROW,
 627        "&&": TokenType.DAMP,
 628        "??": TokenType.DQMARK,
 629        "ALL": TokenType.ALL,
 630        "ALWAYS": TokenType.ALWAYS,
 631        "AND": TokenType.AND,
 632        "ANTI": TokenType.ANTI,
 633        "ANY": TokenType.ANY,
 634        "ASC": TokenType.ASC,
 635        "AS": TokenType.ALIAS,
 636        "ASOF": TokenType.ASOF,
 637        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 638        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 639        "BEGIN": TokenType.BEGIN,
 640        "BETWEEN": TokenType.BETWEEN,
 641        "CACHE": TokenType.CACHE,
 642        "UNCACHE": TokenType.UNCACHE,
 643        "CASE": TokenType.CASE,
 644        "CHARACTER SET": TokenType.CHARACTER_SET,
 645        "CLUSTER BY": TokenType.CLUSTER_BY,
 646        "COLLATE": TokenType.COLLATE,
 647        "COLUMN": TokenType.COLUMN,
 648        "COMMIT": TokenType.COMMIT,
 649        "CONNECT BY": TokenType.CONNECT_BY,
 650        "CONSTRAINT": TokenType.CONSTRAINT,
 651        "COPY": TokenType.COPY,
 652        "CREATE": TokenType.CREATE,
 653        "CROSS": TokenType.CROSS,
 654        "CUBE": TokenType.CUBE,
 655        "CURRENT_DATE": TokenType.CURRENT_DATE,
 656        "CURRENT_TIME": TokenType.CURRENT_TIME,
 657        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 658        "CURRENT_USER": TokenType.CURRENT_USER,
 659        "DATABASE": TokenType.DATABASE,
 660        "DEFAULT": TokenType.DEFAULT,
 661        "DELETE": TokenType.DELETE,
 662        "DESC": TokenType.DESC,
 663        "DESCRIBE": TokenType.DESCRIBE,
 664        "DISTINCT": TokenType.DISTINCT,
 665        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 666        "DIV": TokenType.DIV,
 667        "DROP": TokenType.DROP,
 668        "ELSE": TokenType.ELSE,
 669        "END": TokenType.END,
 670        "ENUM": TokenType.ENUM,
 671        "ESCAPE": TokenType.ESCAPE,
 672        "EXCEPT": TokenType.EXCEPT,
 673        "EXECUTE": TokenType.EXECUTE,
 674        "EXISTS": TokenType.EXISTS,
 675        "FALSE": TokenType.FALSE,
 676        "FETCH": TokenType.FETCH,
 677        "FILTER": TokenType.FILTER,
 678        "FIRST": TokenType.FIRST,
 679        "FULL": TokenType.FULL,
 680        "FUNCTION": TokenType.FUNCTION,
 681        "FOR": TokenType.FOR,
 682        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 683        "FORMAT": TokenType.FORMAT,
 684        "FROM": TokenType.FROM,
 685        "GEOGRAPHY": TokenType.GEOGRAPHY,
 686        "GEOMETRY": TokenType.GEOMETRY,
 687        "GLOB": TokenType.GLOB,
 688        "GROUP BY": TokenType.GROUP_BY,
 689        "GROUPING SETS": TokenType.GROUPING_SETS,
 690        "HAVING": TokenType.HAVING,
 691        "ILIKE": TokenType.ILIKE,
 692        "IN": TokenType.IN,
 693        "INDEX": TokenType.INDEX,
 694        "INET": TokenType.INET,
 695        "INNER": TokenType.INNER,
 696        "INSERT": TokenType.INSERT,
 697        "INTERVAL": TokenType.INTERVAL,
 698        "INTERSECT": TokenType.INTERSECT,
 699        "INTO": TokenType.INTO,
 700        "IS": TokenType.IS,
 701        "ISNULL": TokenType.ISNULL,
 702        "JOIN": TokenType.JOIN,
 703        "KEEP": TokenType.KEEP,
 704        "KILL": TokenType.KILL,
 705        "LATERAL": TokenType.LATERAL,
 706        "LEFT": TokenType.LEFT,
 707        "LIKE": TokenType.LIKE,
 708        "LIMIT": TokenType.LIMIT,
 709        "LOAD": TokenType.LOAD,
 710        "LOCK": TokenType.LOCK,
 711        "MERGE": TokenType.MERGE,
 712        "NATURAL": TokenType.NATURAL,
 713        "NEXT": TokenType.NEXT,
 714        "NOT": TokenType.NOT,
 715        "NOTNULL": TokenType.NOTNULL,
 716        "NULL": TokenType.NULL,
 717        "OBJECT": TokenType.OBJECT,
 718        "OFFSET": TokenType.OFFSET,
 719        "ON": TokenType.ON,
 720        "OR": TokenType.OR,
 721        "XOR": TokenType.XOR,
 722        "ORDER BY": TokenType.ORDER_BY,
 723        "ORDINALITY": TokenType.ORDINALITY,
 724        "OUTER": TokenType.OUTER,
 725        "OVER": TokenType.OVER,
 726        "OVERLAPS": TokenType.OVERLAPS,
 727        "OVERWRITE": TokenType.OVERWRITE,
 728        "PARTITION": TokenType.PARTITION,
 729        "PARTITION BY": TokenType.PARTITION_BY,
 730        "PARTITIONED BY": TokenType.PARTITION_BY,
 731        "PARTITIONED_BY": TokenType.PARTITION_BY,
 732        "PERCENT": TokenType.PERCENT,
 733        "PIVOT": TokenType.PIVOT,
 734        "PRAGMA": TokenType.PRAGMA,
 735        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 736        "PROCEDURE": TokenType.PROCEDURE,
 737        "QUALIFY": TokenType.QUALIFY,
 738        "RANGE": TokenType.RANGE,
 739        "RECURSIVE": TokenType.RECURSIVE,
 740        "REGEXP": TokenType.RLIKE,
 741        "REPLACE": TokenType.REPLACE,
 742        "RETURNING": TokenType.RETURNING,
 743        "REFERENCES": TokenType.REFERENCES,
 744        "RIGHT": TokenType.RIGHT,
 745        "RLIKE": TokenType.RLIKE,
 746        "ROLLBACK": TokenType.ROLLBACK,
 747        "ROLLUP": TokenType.ROLLUP,
 748        "ROW": TokenType.ROW,
 749        "ROWS": TokenType.ROWS,
 750        "SCHEMA": TokenType.SCHEMA,
 751        "SELECT": TokenType.SELECT,
 752        "SEMI": TokenType.SEMI,
 753        "SET": TokenType.SET,
 754        "SETTINGS": TokenType.SETTINGS,
 755        "SHOW": TokenType.SHOW,
 756        "SIMILAR TO": TokenType.SIMILAR_TO,
 757        "SOME": TokenType.SOME,
 758        "SORT BY": TokenType.SORT_BY,
 759        "START WITH": TokenType.START_WITH,
 760        "TABLE": TokenType.TABLE,
 761        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 762        "TEMP": TokenType.TEMPORARY,
 763        "TEMPORARY": TokenType.TEMPORARY,
 764        "THEN": TokenType.THEN,
 765        "TRUE": TokenType.TRUE,
 766        "TRUNCATE": TokenType.TRUNCATE,
 767        "UNION": TokenType.UNION,
 768        "UNKNOWN": TokenType.UNKNOWN,
 769        "UNNEST": TokenType.UNNEST,
 770        "UNPIVOT": TokenType.UNPIVOT,
 771        "UPDATE": TokenType.UPDATE,
 772        "USE": TokenType.USE,
 773        "USING": TokenType.USING,
 774        "UUID": TokenType.UUID,
 775        "VALUES": TokenType.VALUES,
 776        "VIEW": TokenType.VIEW,
 777        "VOLATILE": TokenType.VOLATILE,
 778        "WHEN": TokenType.WHEN,
 779        "WHERE": TokenType.WHERE,
 780        "WINDOW": TokenType.WINDOW,
 781        "WITH": TokenType.WITH,
 782        "APPLY": TokenType.APPLY,
 783        "ARRAY": TokenType.ARRAY,
 784        "BIT": TokenType.BIT,
 785        "BOOL": TokenType.BOOLEAN,
 786        "BOOLEAN": TokenType.BOOLEAN,
 787        "BYTE": TokenType.TINYINT,
 788        "MEDIUMINT": TokenType.MEDIUMINT,
 789        "INT1": TokenType.TINYINT,
 790        "TINYINT": TokenType.TINYINT,
 791        "INT16": TokenType.SMALLINT,
 792        "SHORT": TokenType.SMALLINT,
 793        "SMALLINT": TokenType.SMALLINT,
 794        "INT128": TokenType.INT128,
 795        "HUGEINT": TokenType.INT128,
 796        "INT2": TokenType.SMALLINT,
 797        "INTEGER": TokenType.INT,
 798        "INT": TokenType.INT,
 799        "INT4": TokenType.INT,
 800        "INT32": TokenType.INT,
 801        "INT64": TokenType.BIGINT,
 802        "LONG": TokenType.BIGINT,
 803        "BIGINT": TokenType.BIGINT,
 804        "INT8": TokenType.TINYINT,
 805        "UINT": TokenType.UINT,
 806        "DEC": TokenType.DECIMAL,
 807        "DECIMAL": TokenType.DECIMAL,
 808        "BIGDECIMAL": TokenType.BIGDECIMAL,
 809        "BIGNUMERIC": TokenType.BIGDECIMAL,
 810        "MAP": TokenType.MAP,
 811        "NULLABLE": TokenType.NULLABLE,
 812        "NUMBER": TokenType.DECIMAL,
 813        "NUMERIC": TokenType.DECIMAL,
 814        "FIXED": TokenType.DECIMAL,
 815        "REAL": TokenType.FLOAT,
 816        "FLOAT": TokenType.FLOAT,
 817        "FLOAT4": TokenType.FLOAT,
 818        "FLOAT8": TokenType.DOUBLE,
 819        "DOUBLE": TokenType.DOUBLE,
 820        "DOUBLE PRECISION": TokenType.DOUBLE,
 821        "JSON": TokenType.JSON,
 822        "CHAR": TokenType.CHAR,
 823        "CHARACTER": TokenType.CHAR,
 824        "NCHAR": TokenType.NCHAR,
 825        "VARCHAR": TokenType.VARCHAR,
 826        "VARCHAR2": TokenType.VARCHAR,
 827        "NVARCHAR": TokenType.NVARCHAR,
 828        "NVARCHAR2": TokenType.NVARCHAR,
 829        "BPCHAR": TokenType.BPCHAR,
 830        "STR": TokenType.TEXT,
 831        "STRING": TokenType.TEXT,
 832        "TEXT": TokenType.TEXT,
 833        "LONGTEXT": TokenType.LONGTEXT,
 834        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 835        "TINYTEXT": TokenType.TINYTEXT,
 836        "CLOB": TokenType.TEXT,
 837        "LONGVARCHAR": TokenType.TEXT,
 838        "BINARY": TokenType.BINARY,
 839        "BLOB": TokenType.VARBINARY,
 840        "LONGBLOB": TokenType.LONGBLOB,
 841        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 842        "TINYBLOB": TokenType.TINYBLOB,
 843        "BYTEA": TokenType.VARBINARY,
 844        "VARBINARY": TokenType.VARBINARY,
 845        "TIME": TokenType.TIME,
 846        "TIMETZ": TokenType.TIMETZ,
 847        "TIMESTAMP": TokenType.TIMESTAMP,
 848        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 849        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 850        "DATE": TokenType.DATE,
 851        "DATETIME": TokenType.DATETIME,
 852        "INT4RANGE": TokenType.INT4RANGE,
 853        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 854        "INT8RANGE": TokenType.INT8RANGE,
 855        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 856        "NUMRANGE": TokenType.NUMRANGE,
 857        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 858        "TSRANGE": TokenType.TSRANGE,
 859        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 860        "TSTZRANGE": TokenType.TSTZRANGE,
 861        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 862        "DATERANGE": TokenType.DATERANGE,
 863        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 864        "UNIQUE": TokenType.UNIQUE,
 865        "STRUCT": TokenType.STRUCT,
 866        "SEQUENCE": TokenType.SEQUENCE,
 867        "VARIANT": TokenType.VARIANT,
 868        "ALTER": TokenType.ALTER,
 869        "ANALYZE": TokenType.COMMAND,
 870        "CALL": TokenType.COMMAND,
 871        "COMMENT": TokenType.COMMENT,
 872        "EXPLAIN": TokenType.COMMAND,
 873        "GRANT": TokenType.COMMAND,
 874        "OPTIMIZE": TokenType.COMMAND,
 875        "PREPARE": TokenType.COMMAND,
 876        "VACUUM": TokenType.COMMAND,
 877        "USER-DEFINED": TokenType.USERDEFINED,
 878        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 879        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 880    }
 881
 882    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 883        " ": TokenType.SPACE,
 884        "\t": TokenType.SPACE,
 885        "\n": TokenType.BREAK,
 886        "\r": TokenType.BREAK,
 887    }
 888
 889    COMMANDS = {
 890        TokenType.COMMAND,
 891        TokenType.EXECUTE,
 892        TokenType.FETCH,
 893        TokenType.SHOW,
 894    }
 895
 896    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 897
 898    # Handle numeric literals like in hive (3L = BIGINT)
 899    NUMERIC_LITERALS: t.Dict[str, str] = {}
 900
 901    COMMENTS = ["--", ("/*", "*/")]
 902
 903    __slots__ = (
 904        "sql",
 905        "size",
 906        "tokens",
 907        "dialect",
 908        "_start",
 909        "_current",
 910        "_line",
 911        "_col",
 912        "_comments",
 913        "_char",
 914        "_end",
 915        "_peek",
 916        "_prev_token_line",
 917        "_rs_dialect_settings",
 918    )
 919
 920    def __init__(self, dialect: DialectType = None) -> None:
 921        from sqlglot.dialects import Dialect
 922
 923        self.dialect = Dialect.get_or_raise(dialect)
 924
 925        if USE_RS_TOKENIZER:
 926            self._rs_dialect_settings = RsTokenizerDialectSettings(
 927                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 928                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 929            )
 930
 931        self.reset()
 932
 933    def reset(self) -> None:
 934        self.sql = ""
 935        self.size = 0
 936        self.tokens: t.List[Token] = []
 937        self._start = 0
 938        self._current = 0
 939        self._line = 1
 940        self._col = 0
 941        self._comments: t.List[str] = []
 942
 943        self._char = ""
 944        self._end = False
 945        self._peek = ""
 946        self._prev_token_line = -1
 947
 948    def tokenize(self, sql: str) -> t.List[Token]:
 949        """Returns a list of tokens corresponding to the SQL string `sql`."""
 950        if USE_RS_TOKENIZER:
 951            return self.tokenize_rs(sql)
 952
 953        self.reset()
 954        self.sql = sql
 955        self.size = len(sql)
 956
 957        try:
 958            self._scan()
 959        except Exception as e:
 960            start = max(self._current - 50, 0)
 961            end = min(self._current + 50, self.size - 1)
 962            context = self.sql[start:end]
 963            raise TokenError(f"Error tokenizing '{context}'") from e
 964
 965        return self.tokens
 966
 967    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
 968        while self.size and not self._end:
 969            current = self._current
 970
 971            # Skip spaces here rather than iteratively calling advance() for performance reasons
 972            while current < self.size:
 973                char = self.sql[current]
 974
 975                if char.isspace() and (char == " " or char == "\t"):
 976                    current += 1
 977                else:
 978                    break
 979
 980            offset = current - self._current if current > self._current else 1
 981
 982            self._start = current
 983            self._advance(offset)
 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 not (self._char == "\r" and self._peek == "\n"):
1012                self._col = 1
1013                self._line += 1
1014        else:
1015            self._col += i
1016
1017        self._current += i
1018        self._end = self._current >= self.size
1019        self._char = self.sql[self._current - 1]
1020        self._peek = "" if self._end else self.sql[self._current]
1021
1022        if alnum and self._char.isalnum():
1023            # Here we use local variables instead of attributes for better performance
1024            _col = self._col
1025            _current = self._current
1026            _end = self._end
1027            _peek = self._peek
1028
1029            while _peek.isalnum():
1030                _col += 1
1031                _current += 1
1032                _end = _current >= self.size
1033                _peek = "" if _end else self.sql[_current]
1034
1035            self._col = _col
1036            self._current = _current
1037            self._end = _end
1038            self._peek = _peek
1039            self._char = self.sql[_current - 1]
1040
1041    @property
1042    def _text(self) -> str:
1043        return self.sql[self._start : self._current]
1044
1045    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1046        self._prev_token_line = self._line
1047
1048        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1049            self.tokens[-1].comments.extend(self._comments)
1050            self._comments = []
1051
1052        self.tokens.append(
1053            Token(
1054                token_type,
1055                text=self._text if text is None else text,
1056                line=self._line,
1057                col=self._col,
1058                start=self._start,
1059                end=self._current - 1,
1060                comments=self._comments,
1061            )
1062        )
1063        self._comments = []
1064
1065        # If we have either a semicolon or a begin token before the command's token, we'll parse
1066        # whatever follows the command's token as a string
1067        if (
1068            token_type in self.COMMANDS
1069            and self._peek != ";"
1070            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1071        ):
1072            start = self._current
1073            tokens = len(self.tokens)
1074            self._scan(lambda: self._peek == ";")
1075            self.tokens = self.tokens[:tokens]
1076            text = self.sql[start : self._current].strip()
1077            if text:
1078                self._add(TokenType.STRING, text)
1079
1080    def _scan_keywords(self) -> None:
1081        size = 0
1082        word = None
1083        chars = self._text
1084        char = chars
1085        prev_space = False
1086        skip = False
1087        trie = self._KEYWORD_TRIE
1088        single_token = char in self.SINGLE_TOKENS
1089
1090        while chars:
1091            if skip:
1092                result = TrieResult.PREFIX
1093            else:
1094                result, trie = in_trie(trie, char.upper())
1095
1096            if result == TrieResult.FAILED:
1097                break
1098            if result == TrieResult.EXISTS:
1099                word = chars
1100
1101            end = self._current + size
1102            size += 1
1103
1104            if end < self.size:
1105                char = self.sql[end]
1106                single_token = single_token or char in self.SINGLE_TOKENS
1107                is_space = char.isspace()
1108
1109                if not is_space or not prev_space:
1110                    if is_space:
1111                        char = " "
1112                    chars += char
1113                    prev_space = is_space
1114                    skip = False
1115                else:
1116                    skip = True
1117            else:
1118                char = ""
1119                break
1120
1121        if word:
1122            if self._scan_string(word):
1123                return
1124            if self._scan_comment(word):
1125                return
1126            if prev_space or single_token or not char:
1127                self._advance(size - 1)
1128                word = word.upper()
1129                self._add(self.KEYWORDS[word], text=word)
1130                return
1131
1132        if self._char in self.SINGLE_TOKENS:
1133            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1134            return
1135
1136        self._scan_var()
1137
1138    def _scan_comment(self, comment_start: str) -> bool:
1139        if comment_start not in self._COMMENTS:
1140            return False
1141
1142        comment_start_line = self._line
1143        comment_start_size = len(comment_start)
1144        comment_end = self._COMMENTS[comment_start]
1145
1146        if comment_end:
1147            # Skip the comment's start delimiter
1148            self._advance(comment_start_size)
1149
1150            comment_end_size = len(comment_end)
1151            while not self._end and self._chars(comment_end_size) != comment_end:
1152                self._advance(alnum=True)
1153
1154            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1155            self._advance(comment_end_size - 1)
1156        else:
1157            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1158                self._advance(alnum=True)
1159            self._comments.append(self._text[comment_start_size:])
1160
1161        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1162        # Multiple consecutive comments are preserved by appending them to the current comments list.
1163        if comment_start_line == self._prev_token_line:
1164            self.tokens[-1].comments.extend(self._comments)
1165            self._comments = []
1166            self._prev_token_line = self._line
1167
1168        return True
1169
1170    def _scan_number(self) -> None:
1171        if self._char == "0":
1172            peek = self._peek.upper()
1173            if peek == "B":
1174                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1175            elif peek == "X":
1176                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1177
1178        decimal = False
1179        scientific = 0
1180
1181        while True:
1182            if self._peek.isdigit():
1183                self._advance()
1184            elif self._peek == "." and not decimal:
1185                decimal = True
1186                self._advance()
1187            elif self._peek in ("-", "+") and scientific == 1:
1188                scientific += 1
1189                self._advance()
1190            elif self._peek.upper() == "E" and not scientific:
1191                scientific += 1
1192                self._advance()
1193            elif self._peek.isidentifier():
1194                number_text = self._text
1195                literal = ""
1196
1197                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1198                    literal += self._peek
1199                    self._advance()
1200
1201                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1202
1203                if token_type:
1204                    self._add(TokenType.NUMBER, number_text)
1205                    self._add(TokenType.DCOLON, "::")
1206                    return self._add(token_type, literal)
1207                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1208                    return self._add(TokenType.VAR)
1209
1210                self._advance(-len(literal))
1211                return self._add(TokenType.NUMBER, number_text)
1212            else:
1213                return self._add(TokenType.NUMBER)
1214
1215    def _scan_bits(self) -> None:
1216        self._advance()
1217        value = self._extract_value()
1218        try:
1219            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1220            int(value, 2)
1221            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1222        except ValueError:
1223            self._add(TokenType.IDENTIFIER)
1224
1225    def _scan_hex(self) -> None:
1226        self._advance()
1227        value = self._extract_value()
1228        try:
1229            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1230            int(value, 16)
1231            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1232        except ValueError:
1233            self._add(TokenType.IDENTIFIER)
1234
1235    def _extract_value(self) -> str:
1236        while True:
1237            char = self._peek.strip()
1238            if char and char not in self.SINGLE_TOKENS:
1239                self._advance(alnum=True)
1240            else:
1241                break
1242
1243        return self._text
1244
1245    def _scan_string(self, start: str) -> bool:
1246        base = None
1247        token_type = TokenType.STRING
1248
1249        if start in self._QUOTES:
1250            end = self._QUOTES[start]
1251        elif start in self._FORMAT_STRINGS:
1252            end, token_type = self._FORMAT_STRINGS[start]
1253
1254            if token_type == TokenType.HEX_STRING:
1255                base = 16
1256            elif token_type == TokenType.BIT_STRING:
1257                base = 2
1258            elif token_type == TokenType.HEREDOC_STRING:
1259                if (
1260                    self.HEREDOC_TAG_IS_IDENTIFIER
1261                    and not self._peek.isidentifier()
1262                    and not self._peek == end
1263                ):
1264                    if self.HEREDOC_STRING_ALTERNATIVE != token_type.VAR:
1265                        self._add(self.HEREDOC_STRING_ALTERNATIVE)
1266                    else:
1267                        self._scan_var()
1268
1269                    return True
1270
1271                self._advance()
1272
1273                if self._char == end:
1274                    tag = ""
1275                else:
1276                    tag = self._extract_string(
1277                        end,
1278                        unescape_sequences=False,
1279                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1280                    )
1281
1282                if self._end and tag and self.HEREDOC_TAG_IS_IDENTIFIER:
1283                    self._advance(-len(tag))
1284                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1285                    return True
1286
1287                end = f"{start}{tag}{end}"
1288        else:
1289            return False
1290
1291        self._advance(len(start))
1292        text = self._extract_string(end, unescape_sequences=token_type != TokenType.RAW_STRING)
1293
1294        if base:
1295            try:
1296                int(text, base)
1297            except Exception:
1298                raise TokenError(
1299                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1300                )
1301
1302        self._add(token_type, text)
1303        return True
1304
1305    def _scan_identifier(self, identifier_end: str) -> None:
1306        self._advance()
1307        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1308        self._add(TokenType.IDENTIFIER, text)
1309
1310    def _scan_var(self) -> None:
1311        while True:
1312            char = self._peek.strip()
1313            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1314                self._advance(alnum=True)
1315            else:
1316                break
1317
1318        self._add(
1319            TokenType.VAR
1320            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1321            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1322        )
1323
1324    def _extract_string(
1325        self,
1326        delimiter: str,
1327        escapes: t.Optional[t.Set[str]] = None,
1328        unescape_sequences: bool = True,
1329        raise_unmatched: bool = True,
1330    ) -> str:
1331        text = ""
1332        delim_size = len(delimiter)
1333        escapes = self._STRING_ESCAPES if escapes is None else escapes
1334
1335        while True:
1336            if (
1337                unescape_sequences
1338                and self.dialect.UNESCAPED_SEQUENCES
1339                and self._peek
1340                and self._char in self.STRING_ESCAPES
1341            ):
1342                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1343                if unescaped_sequence:
1344                    self._advance(2)
1345                    text += unescaped_sequence
1346                    continue
1347            if (
1348                self._char in escapes
1349                and (self._peek == delimiter or self._peek in escapes)
1350                and (self._char not in self._QUOTES or self._char == self._peek)
1351            ):
1352                if self._peek == delimiter:
1353                    text += self._peek
1354                else:
1355                    text += self._char + self._peek
1356
1357                if self._current + 1 < self.size:
1358                    self._advance(2)
1359                else:
1360                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1361            else:
1362                if self._chars(delim_size) == delimiter:
1363                    if delim_size > 1:
1364                        self._advance(delim_size - 1)
1365                    break
1366
1367                if self._end:
1368                    if not raise_unmatched:
1369                        return text + self._char
1370
1371                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1372
1373                current = self._current - 1
1374                self._advance(alnum=True)
1375                text += self.sql[current : self._current - 1]
1376
1377        return text
1378
1379    def tokenize_rs(self, sql: str) -> t.List[Token]:
1380        if not self._RS_TOKENIZER:
1381            raise SqlglotError("Rust tokenizer is not available")
1382
1383        try:
1384            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1385            for token in tokens:
1386                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1387            return tokens
1388        except Exception as e:
1389            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    COPY = auto()
228    CREATE = auto()
229    CROSS = auto()
230    CUBE = auto()
231    CURRENT_DATE = auto()
232    CURRENT_DATETIME = auto()
233    CURRENT_TIME = auto()
234    CURRENT_TIMESTAMP = auto()
235    CURRENT_USER = auto()
236    DEFAULT = auto()
237    DELETE = auto()
238    DESC = auto()
239    DESCRIBE = auto()
240    DICTIONARY = auto()
241    DISTINCT = auto()
242    DISTRIBUTE_BY = auto()
243    DIV = auto()
244    DROP = auto()
245    ELSE = auto()
246    END = auto()
247    ESCAPE = auto()
248    EXCEPT = auto()
249    EXECUTE = auto()
250    EXISTS = auto()
251    FALSE = auto()
252    FETCH = auto()
253    FILTER = auto()
254    FINAL = auto()
255    FIRST = auto()
256    FOR = auto()
257    FORCE = auto()
258    FOREIGN_KEY = auto()
259    FORMAT = auto()
260    FROM = auto()
261    FULL = auto()
262    FUNCTION = auto()
263    GLOB = auto()
264    GLOBAL = auto()
265    GROUP_BY = auto()
266    GROUPING_SETS = auto()
267    HAVING = auto()
268    HINT = auto()
269    IGNORE = auto()
270    ILIKE = auto()
271    ILIKE_ANY = auto()
272    IN = auto()
273    INDEX = auto()
274    INNER = auto()
275    INSERT = auto()
276    INTERSECT = auto()
277    INTERVAL = auto()
278    INTO = auto()
279    INTRODUCER = auto()
280    IRLIKE = auto()
281    IS = auto()
282    ISNULL = auto()
283    JOIN = auto()
284    JOIN_MARKER = auto()
285    KEEP = auto()
286    KILL = auto()
287    LANGUAGE = auto()
288    LATERAL = auto()
289    LEFT = auto()
290    LIKE = auto()
291    LIKE_ANY = auto()
292    LIMIT = auto()
293    LOAD = auto()
294    LOCK = auto()
295    MAP = auto()
296    MATCH_CONDITION = auto()
297    MATCH_RECOGNIZE = auto()
298    MEMBER_OF = auto()
299    MERGE = auto()
300    MOD = auto()
301    MODEL = auto()
302    NATURAL = auto()
303    NEXT = auto()
304    NOTNULL = auto()
305    NULL = auto()
306    OBJECT_IDENTIFIER = auto()
307    OFFSET = auto()
308    ON = auto()
309    ONLY = auto()
310    OPERATOR = auto()
311    ORDER_BY = auto()
312    ORDER_SIBLINGS_BY = auto()
313    ORDERED = auto()
314    ORDINALITY = auto()
315    OUTER = auto()
316    OVER = auto()
317    OVERLAPS = auto()
318    OVERWRITE = auto()
319    PARTITION = auto()
320    PARTITION_BY = auto()
321    PERCENT = auto()
322    PIVOT = auto()
323    PLACEHOLDER = auto()
324    POSITIONAL = auto()
325    PRAGMA = auto()
326    PREWHERE = auto()
327    PRIMARY_KEY = auto()
328    PROCEDURE = auto()
329    PROPERTIES = auto()
330    PSEUDO_TYPE = auto()
331    QUALIFY = auto()
332    QUOTE = auto()
333    RANGE = auto()
334    RECURSIVE = auto()
335    REFRESH = auto()
336    REPLACE = auto()
337    RETURNING = auto()
338    REFERENCES = auto()
339    RIGHT = auto()
340    RLIKE = auto()
341    ROLLBACK = auto()
342    ROLLUP = auto()
343    ROW = auto()
344    ROWS = auto()
345    SELECT = auto()
346    SEMI = auto()
347    SEPARATOR = auto()
348    SEQUENCE = auto()
349    SERDE_PROPERTIES = auto()
350    SET = auto()
351    SETTINGS = auto()
352    SHOW = auto()
353    SIMILAR_TO = auto()
354    SOME = auto()
355    SORT_BY = auto()
356    START_WITH = auto()
357    STORAGE_INTEGRATION = auto()
358    STRUCT = auto()
359    TABLE_SAMPLE = auto()
360    TEMPORARY = auto()
361    TOP = auto()
362    THEN = auto()
363    TRUE = auto()
364    TRUNCATE = auto()
365    UNCACHE = auto()
366    UNION = auto()
367    UNNEST = auto()
368    UNPIVOT = auto()
369    UPDATE = auto()
370    USE = auto()
371    USING = auto()
372    VALUES = auto()
373    VIEW = auto()
374    VOLATILE = auto()
375    WHEN = auto()
376    WHERE = auto()
377    WINDOW = auto()
378    WITH = auto()
379    UNIQUE = auto()
380    VERSION_SNAPSHOT = auto()
381    TIMESTAMP_SNAPSHOT = auto()
382    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'>
COPY = <TokenType.COPY: 'COPY'>
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_CONDITION = <TokenType.MATCH_CONDITION: 'MATCH_CONDITION'>
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:
389class Token:
390    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
391
392    @classmethod
393    def number(cls, number: int) -> Token:
394        """Returns a NUMBER token with `number` as its text."""
395        return cls(TokenType.NUMBER, str(number))
396
397    @classmethod
398    def string(cls, string: str) -> Token:
399        """Returns a STRING token with `string` as its text."""
400        return cls(TokenType.STRING, string)
401
402    @classmethod
403    def identifier(cls, identifier: str) -> Token:
404        """Returns an IDENTIFIER token with `identifier` as its text."""
405        return cls(TokenType.IDENTIFIER, identifier)
406
407    @classmethod
408    def var(cls, var: str) -> Token:
409        """Returns an VAR token with `var` as its text."""
410        return cls(TokenType.VAR, var)
411
412    def __init__(
413        self,
414        token_type: TokenType,
415        text: str,
416        line: int = 1,
417        col: int = 1,
418        start: int = 0,
419        end: int = 0,
420        comments: t.Optional[t.List[str]] = None,
421    ) -> None:
422        """Token initializer.
423
424        Args:
425            token_type: The TokenType Enum.
426            text: The text of the token.
427            line: The line that the token ends on.
428            col: The column that the token ends on.
429            start: The start index of the token.
430            end: The ending index of the token.
431            comments: The comments to attach to the token.
432        """
433        self.token_type = token_type
434        self.text = text
435        self.line = line
436        self.col = col
437        self.start = start
438        self.end = end
439        self.comments = [] if comments is None else comments
440
441    def __repr__(self) -> str:
442        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
443        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)
412    def __init__(
413        self,
414        token_type: TokenType,
415        text: str,
416        line: int = 1,
417        col: int = 1,
418        start: int = 0,
419        end: int = 0,
420        comments: t.Optional[t.List[str]] = None,
421    ) -> None:
422        """Token initializer.
423
424        Args:
425            token_type: The TokenType Enum.
426            text: The text of the token.
427            line: The line that the token ends on.
428            col: The column that the token ends on.
429            start: The start index of the token.
430            end: The ending index of the token.
431            comments: The comments to attach to the token.
432        """
433        self.token_type = token_type
434        self.text = text
435        self.line = line
436        self.col = col
437        self.start = start
438        self.end = end
439        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:
392    @classmethod
393    def number(cls, number: int) -> Token:
394        """Returns a NUMBER token with `number` as its text."""
395        return cls(TokenType.NUMBER, str(number))

Returns a NUMBER token with number as its text.

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

Returns a STRING token with string as its text.

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

Returns an IDENTIFIER token with identifier as its text.

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

Returns an VAR token with var as its text.

token_type
text
line
col
start
end
comments
class Tokenizer:
 544class Tokenizer(metaclass=_Tokenizer):
 545    SINGLE_TOKENS = {
 546        "(": TokenType.L_PAREN,
 547        ")": TokenType.R_PAREN,
 548        "[": TokenType.L_BRACKET,
 549        "]": TokenType.R_BRACKET,
 550        "{": TokenType.L_BRACE,
 551        "}": TokenType.R_BRACE,
 552        "&": TokenType.AMP,
 553        "^": TokenType.CARET,
 554        ":": TokenType.COLON,
 555        ",": TokenType.COMMA,
 556        ".": TokenType.DOT,
 557        "-": TokenType.DASH,
 558        "=": TokenType.EQ,
 559        ">": TokenType.GT,
 560        "<": TokenType.LT,
 561        "%": TokenType.MOD,
 562        "!": TokenType.NOT,
 563        "|": TokenType.PIPE,
 564        "+": TokenType.PLUS,
 565        ";": TokenType.SEMICOLON,
 566        "/": TokenType.SLASH,
 567        "\\": TokenType.BACKSLASH,
 568        "*": TokenType.STAR,
 569        "~": TokenType.TILDA,
 570        "?": TokenType.PLACEHOLDER,
 571        "@": TokenType.PARAMETER,
 572        "#": TokenType.HASH,
 573        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 574        "'": TokenType.UNKNOWN,
 575        "`": TokenType.UNKNOWN,
 576        '"': TokenType.UNKNOWN,
 577    }
 578
 579    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 580    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 581    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 582    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 583    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 584    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 585    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 586    IDENTIFIER_ESCAPES = ['"']
 587    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 588    STRING_ESCAPES = ["'"]
 589    VAR_SINGLE_TOKENS: t.Set[str] = set()
 590
 591    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 592    HEREDOC_TAG_IS_IDENTIFIER = False
 593
 594    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 595    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 596
 597    # Autofilled
 598    _COMMENTS: t.Dict[str, str] = {}
 599    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 600    _IDENTIFIERS: t.Dict[str, str] = {}
 601    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 602    _QUOTES: t.Dict[str, str] = {}
 603    _STRING_ESCAPES: t.Set[str] = set()
 604    _KEYWORD_TRIE: t.Dict = {}
 605    _RS_TOKENIZER: t.Optional[t.Any] = None
 606
 607    KEYWORDS: t.Dict[str, TokenType] = {
 608        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 609        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 610        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 611        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 612        "/*+": TokenType.HINT,
 613        "==": TokenType.EQ,
 614        "::": TokenType.DCOLON,
 615        "||": TokenType.DPIPE,
 616        ">=": TokenType.GTE,
 617        "<=": TokenType.LTE,
 618        "<>": TokenType.NEQ,
 619        "!=": TokenType.NEQ,
 620        ":=": TokenType.COLON_EQ,
 621        "<=>": TokenType.NULLSAFE_EQ,
 622        "->": TokenType.ARROW,
 623        "->>": TokenType.DARROW,
 624        "=>": TokenType.FARROW,
 625        "#>": TokenType.HASH_ARROW,
 626        "#>>": TokenType.DHASH_ARROW,
 627        "<->": TokenType.LR_ARROW,
 628        "&&": TokenType.DAMP,
 629        "??": TokenType.DQMARK,
 630        "ALL": TokenType.ALL,
 631        "ALWAYS": TokenType.ALWAYS,
 632        "AND": TokenType.AND,
 633        "ANTI": TokenType.ANTI,
 634        "ANY": TokenType.ANY,
 635        "ASC": TokenType.ASC,
 636        "AS": TokenType.ALIAS,
 637        "ASOF": TokenType.ASOF,
 638        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 639        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 640        "BEGIN": TokenType.BEGIN,
 641        "BETWEEN": TokenType.BETWEEN,
 642        "CACHE": TokenType.CACHE,
 643        "UNCACHE": TokenType.UNCACHE,
 644        "CASE": TokenType.CASE,
 645        "CHARACTER SET": TokenType.CHARACTER_SET,
 646        "CLUSTER BY": TokenType.CLUSTER_BY,
 647        "COLLATE": TokenType.COLLATE,
 648        "COLUMN": TokenType.COLUMN,
 649        "COMMIT": TokenType.COMMIT,
 650        "CONNECT BY": TokenType.CONNECT_BY,
 651        "CONSTRAINT": TokenType.CONSTRAINT,
 652        "COPY": TokenType.COPY,
 653        "CREATE": TokenType.CREATE,
 654        "CROSS": TokenType.CROSS,
 655        "CUBE": TokenType.CUBE,
 656        "CURRENT_DATE": TokenType.CURRENT_DATE,
 657        "CURRENT_TIME": TokenType.CURRENT_TIME,
 658        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 659        "CURRENT_USER": TokenType.CURRENT_USER,
 660        "DATABASE": TokenType.DATABASE,
 661        "DEFAULT": TokenType.DEFAULT,
 662        "DELETE": TokenType.DELETE,
 663        "DESC": TokenType.DESC,
 664        "DESCRIBE": TokenType.DESCRIBE,
 665        "DISTINCT": TokenType.DISTINCT,
 666        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 667        "DIV": TokenType.DIV,
 668        "DROP": TokenType.DROP,
 669        "ELSE": TokenType.ELSE,
 670        "END": TokenType.END,
 671        "ENUM": TokenType.ENUM,
 672        "ESCAPE": TokenType.ESCAPE,
 673        "EXCEPT": TokenType.EXCEPT,
 674        "EXECUTE": TokenType.EXECUTE,
 675        "EXISTS": TokenType.EXISTS,
 676        "FALSE": TokenType.FALSE,
 677        "FETCH": TokenType.FETCH,
 678        "FILTER": TokenType.FILTER,
 679        "FIRST": TokenType.FIRST,
 680        "FULL": TokenType.FULL,
 681        "FUNCTION": TokenType.FUNCTION,
 682        "FOR": TokenType.FOR,
 683        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 684        "FORMAT": TokenType.FORMAT,
 685        "FROM": TokenType.FROM,
 686        "GEOGRAPHY": TokenType.GEOGRAPHY,
 687        "GEOMETRY": TokenType.GEOMETRY,
 688        "GLOB": TokenType.GLOB,
 689        "GROUP BY": TokenType.GROUP_BY,
 690        "GROUPING SETS": TokenType.GROUPING_SETS,
 691        "HAVING": TokenType.HAVING,
 692        "ILIKE": TokenType.ILIKE,
 693        "IN": TokenType.IN,
 694        "INDEX": TokenType.INDEX,
 695        "INET": TokenType.INET,
 696        "INNER": TokenType.INNER,
 697        "INSERT": TokenType.INSERT,
 698        "INTERVAL": TokenType.INTERVAL,
 699        "INTERSECT": TokenType.INTERSECT,
 700        "INTO": TokenType.INTO,
 701        "IS": TokenType.IS,
 702        "ISNULL": TokenType.ISNULL,
 703        "JOIN": TokenType.JOIN,
 704        "KEEP": TokenType.KEEP,
 705        "KILL": TokenType.KILL,
 706        "LATERAL": TokenType.LATERAL,
 707        "LEFT": TokenType.LEFT,
 708        "LIKE": TokenType.LIKE,
 709        "LIMIT": TokenType.LIMIT,
 710        "LOAD": TokenType.LOAD,
 711        "LOCK": TokenType.LOCK,
 712        "MERGE": TokenType.MERGE,
 713        "NATURAL": TokenType.NATURAL,
 714        "NEXT": TokenType.NEXT,
 715        "NOT": TokenType.NOT,
 716        "NOTNULL": TokenType.NOTNULL,
 717        "NULL": TokenType.NULL,
 718        "OBJECT": TokenType.OBJECT,
 719        "OFFSET": TokenType.OFFSET,
 720        "ON": TokenType.ON,
 721        "OR": TokenType.OR,
 722        "XOR": TokenType.XOR,
 723        "ORDER BY": TokenType.ORDER_BY,
 724        "ORDINALITY": TokenType.ORDINALITY,
 725        "OUTER": TokenType.OUTER,
 726        "OVER": TokenType.OVER,
 727        "OVERLAPS": TokenType.OVERLAPS,
 728        "OVERWRITE": TokenType.OVERWRITE,
 729        "PARTITION": TokenType.PARTITION,
 730        "PARTITION BY": TokenType.PARTITION_BY,
 731        "PARTITIONED BY": TokenType.PARTITION_BY,
 732        "PARTITIONED_BY": TokenType.PARTITION_BY,
 733        "PERCENT": TokenType.PERCENT,
 734        "PIVOT": TokenType.PIVOT,
 735        "PRAGMA": TokenType.PRAGMA,
 736        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 737        "PROCEDURE": TokenType.PROCEDURE,
 738        "QUALIFY": TokenType.QUALIFY,
 739        "RANGE": TokenType.RANGE,
 740        "RECURSIVE": TokenType.RECURSIVE,
 741        "REGEXP": TokenType.RLIKE,
 742        "REPLACE": TokenType.REPLACE,
 743        "RETURNING": TokenType.RETURNING,
 744        "REFERENCES": TokenType.REFERENCES,
 745        "RIGHT": TokenType.RIGHT,
 746        "RLIKE": TokenType.RLIKE,
 747        "ROLLBACK": TokenType.ROLLBACK,
 748        "ROLLUP": TokenType.ROLLUP,
 749        "ROW": TokenType.ROW,
 750        "ROWS": TokenType.ROWS,
 751        "SCHEMA": TokenType.SCHEMA,
 752        "SELECT": TokenType.SELECT,
 753        "SEMI": TokenType.SEMI,
 754        "SET": TokenType.SET,
 755        "SETTINGS": TokenType.SETTINGS,
 756        "SHOW": TokenType.SHOW,
 757        "SIMILAR TO": TokenType.SIMILAR_TO,
 758        "SOME": TokenType.SOME,
 759        "SORT BY": TokenType.SORT_BY,
 760        "START WITH": TokenType.START_WITH,
 761        "TABLE": TokenType.TABLE,
 762        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 763        "TEMP": TokenType.TEMPORARY,
 764        "TEMPORARY": TokenType.TEMPORARY,
 765        "THEN": TokenType.THEN,
 766        "TRUE": TokenType.TRUE,
 767        "TRUNCATE": TokenType.TRUNCATE,
 768        "UNION": TokenType.UNION,
 769        "UNKNOWN": TokenType.UNKNOWN,
 770        "UNNEST": TokenType.UNNEST,
 771        "UNPIVOT": TokenType.UNPIVOT,
 772        "UPDATE": TokenType.UPDATE,
 773        "USE": TokenType.USE,
 774        "USING": TokenType.USING,
 775        "UUID": TokenType.UUID,
 776        "VALUES": TokenType.VALUES,
 777        "VIEW": TokenType.VIEW,
 778        "VOLATILE": TokenType.VOLATILE,
 779        "WHEN": TokenType.WHEN,
 780        "WHERE": TokenType.WHERE,
 781        "WINDOW": TokenType.WINDOW,
 782        "WITH": TokenType.WITH,
 783        "APPLY": TokenType.APPLY,
 784        "ARRAY": TokenType.ARRAY,
 785        "BIT": TokenType.BIT,
 786        "BOOL": TokenType.BOOLEAN,
 787        "BOOLEAN": TokenType.BOOLEAN,
 788        "BYTE": TokenType.TINYINT,
 789        "MEDIUMINT": TokenType.MEDIUMINT,
 790        "INT1": TokenType.TINYINT,
 791        "TINYINT": TokenType.TINYINT,
 792        "INT16": TokenType.SMALLINT,
 793        "SHORT": TokenType.SMALLINT,
 794        "SMALLINT": TokenType.SMALLINT,
 795        "INT128": TokenType.INT128,
 796        "HUGEINT": TokenType.INT128,
 797        "INT2": TokenType.SMALLINT,
 798        "INTEGER": TokenType.INT,
 799        "INT": TokenType.INT,
 800        "INT4": TokenType.INT,
 801        "INT32": TokenType.INT,
 802        "INT64": TokenType.BIGINT,
 803        "LONG": TokenType.BIGINT,
 804        "BIGINT": TokenType.BIGINT,
 805        "INT8": TokenType.TINYINT,
 806        "UINT": TokenType.UINT,
 807        "DEC": TokenType.DECIMAL,
 808        "DECIMAL": TokenType.DECIMAL,
 809        "BIGDECIMAL": TokenType.BIGDECIMAL,
 810        "BIGNUMERIC": TokenType.BIGDECIMAL,
 811        "MAP": TokenType.MAP,
 812        "NULLABLE": TokenType.NULLABLE,
 813        "NUMBER": TokenType.DECIMAL,
 814        "NUMERIC": TokenType.DECIMAL,
 815        "FIXED": TokenType.DECIMAL,
 816        "REAL": TokenType.FLOAT,
 817        "FLOAT": TokenType.FLOAT,
 818        "FLOAT4": TokenType.FLOAT,
 819        "FLOAT8": TokenType.DOUBLE,
 820        "DOUBLE": TokenType.DOUBLE,
 821        "DOUBLE PRECISION": TokenType.DOUBLE,
 822        "JSON": TokenType.JSON,
 823        "CHAR": TokenType.CHAR,
 824        "CHARACTER": TokenType.CHAR,
 825        "NCHAR": TokenType.NCHAR,
 826        "VARCHAR": TokenType.VARCHAR,
 827        "VARCHAR2": TokenType.VARCHAR,
 828        "NVARCHAR": TokenType.NVARCHAR,
 829        "NVARCHAR2": TokenType.NVARCHAR,
 830        "BPCHAR": TokenType.BPCHAR,
 831        "STR": TokenType.TEXT,
 832        "STRING": TokenType.TEXT,
 833        "TEXT": TokenType.TEXT,
 834        "LONGTEXT": TokenType.LONGTEXT,
 835        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 836        "TINYTEXT": TokenType.TINYTEXT,
 837        "CLOB": TokenType.TEXT,
 838        "LONGVARCHAR": TokenType.TEXT,
 839        "BINARY": TokenType.BINARY,
 840        "BLOB": TokenType.VARBINARY,
 841        "LONGBLOB": TokenType.LONGBLOB,
 842        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 843        "TINYBLOB": TokenType.TINYBLOB,
 844        "BYTEA": TokenType.VARBINARY,
 845        "VARBINARY": TokenType.VARBINARY,
 846        "TIME": TokenType.TIME,
 847        "TIMETZ": TokenType.TIMETZ,
 848        "TIMESTAMP": TokenType.TIMESTAMP,
 849        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 850        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 851        "DATE": TokenType.DATE,
 852        "DATETIME": TokenType.DATETIME,
 853        "INT4RANGE": TokenType.INT4RANGE,
 854        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 855        "INT8RANGE": TokenType.INT8RANGE,
 856        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 857        "NUMRANGE": TokenType.NUMRANGE,
 858        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 859        "TSRANGE": TokenType.TSRANGE,
 860        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 861        "TSTZRANGE": TokenType.TSTZRANGE,
 862        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 863        "DATERANGE": TokenType.DATERANGE,
 864        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 865        "UNIQUE": TokenType.UNIQUE,
 866        "STRUCT": TokenType.STRUCT,
 867        "SEQUENCE": TokenType.SEQUENCE,
 868        "VARIANT": TokenType.VARIANT,
 869        "ALTER": TokenType.ALTER,
 870        "ANALYZE": TokenType.COMMAND,
 871        "CALL": TokenType.COMMAND,
 872        "COMMENT": TokenType.COMMENT,
 873        "EXPLAIN": TokenType.COMMAND,
 874        "GRANT": TokenType.COMMAND,
 875        "OPTIMIZE": TokenType.COMMAND,
 876        "PREPARE": TokenType.COMMAND,
 877        "VACUUM": TokenType.COMMAND,
 878        "USER-DEFINED": TokenType.USERDEFINED,
 879        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 880        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 881    }
 882
 883    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 884        " ": TokenType.SPACE,
 885        "\t": TokenType.SPACE,
 886        "\n": TokenType.BREAK,
 887        "\r": TokenType.BREAK,
 888    }
 889
 890    COMMANDS = {
 891        TokenType.COMMAND,
 892        TokenType.EXECUTE,
 893        TokenType.FETCH,
 894        TokenType.SHOW,
 895    }
 896
 897    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 898
 899    # Handle numeric literals like in hive (3L = BIGINT)
 900    NUMERIC_LITERALS: t.Dict[str, str] = {}
 901
 902    COMMENTS = ["--", ("/*", "*/")]
 903
 904    __slots__ = (
 905        "sql",
 906        "size",
 907        "tokens",
 908        "dialect",
 909        "_start",
 910        "_current",
 911        "_line",
 912        "_col",
 913        "_comments",
 914        "_char",
 915        "_end",
 916        "_peek",
 917        "_prev_token_line",
 918        "_rs_dialect_settings",
 919    )
 920
 921    def __init__(self, dialect: DialectType = None) -> None:
 922        from sqlglot.dialects import Dialect
 923
 924        self.dialect = Dialect.get_or_raise(dialect)
 925
 926        if USE_RS_TOKENIZER:
 927            self._rs_dialect_settings = RsTokenizerDialectSettings(
 928                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 929                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 930            )
 931
 932        self.reset()
 933
 934    def reset(self) -> None:
 935        self.sql = ""
 936        self.size = 0
 937        self.tokens: t.List[Token] = []
 938        self._start = 0
 939        self._current = 0
 940        self._line = 1
 941        self._col = 0
 942        self._comments: t.List[str] = []
 943
 944        self._char = ""
 945        self._end = False
 946        self._peek = ""
 947        self._prev_token_line = -1
 948
 949    def tokenize(self, sql: str) -> t.List[Token]:
 950        """Returns a list of tokens corresponding to the SQL string `sql`."""
 951        if USE_RS_TOKENIZER:
 952            return self.tokenize_rs(sql)
 953
 954        self.reset()
 955        self.sql = sql
 956        self.size = len(sql)
 957
 958        try:
 959            self._scan()
 960        except Exception as e:
 961            start = max(self._current - 50, 0)
 962            end = min(self._current + 50, self.size - 1)
 963            context = self.sql[start:end]
 964            raise TokenError(f"Error tokenizing '{context}'") from e
 965
 966        return self.tokens
 967
 968    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
 969        while self.size and not self._end:
 970            current = self._current
 971
 972            # Skip spaces here rather than iteratively calling advance() for performance reasons
 973            while current < self.size:
 974                char = self.sql[current]
 975
 976                if char.isspace() and (char == " " or char == "\t"):
 977                    current += 1
 978                else:
 979                    break
 980
 981            offset = current - self._current if current > self._current else 1
 982
 983            self._start = current
 984            self._advance(offset)
 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 not (self._char == "\r" and self._peek == "\n"):
1013                self._col = 1
1014                self._line += 1
1015        else:
1016            self._col += i
1017
1018        self._current += i
1019        self._end = self._current >= self.size
1020        self._char = self.sql[self._current - 1]
1021        self._peek = "" if self._end else self.sql[self._current]
1022
1023        if alnum and self._char.isalnum():
1024            # Here we use local variables instead of attributes for better performance
1025            _col = self._col
1026            _current = self._current
1027            _end = self._end
1028            _peek = self._peek
1029
1030            while _peek.isalnum():
1031                _col += 1
1032                _current += 1
1033                _end = _current >= self.size
1034                _peek = "" if _end else self.sql[_current]
1035
1036            self._col = _col
1037            self._current = _current
1038            self._end = _end
1039            self._peek = _peek
1040            self._char = self.sql[_current - 1]
1041
1042    @property
1043    def _text(self) -> str:
1044        return self.sql[self._start : self._current]
1045
1046    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1047        self._prev_token_line = self._line
1048
1049        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1050            self.tokens[-1].comments.extend(self._comments)
1051            self._comments = []
1052
1053        self.tokens.append(
1054            Token(
1055                token_type,
1056                text=self._text if text is None else text,
1057                line=self._line,
1058                col=self._col,
1059                start=self._start,
1060                end=self._current - 1,
1061                comments=self._comments,
1062            )
1063        )
1064        self._comments = []
1065
1066        # If we have either a semicolon or a begin token before the command's token, we'll parse
1067        # whatever follows the command's token as a string
1068        if (
1069            token_type in self.COMMANDS
1070            and self._peek != ";"
1071            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1072        ):
1073            start = self._current
1074            tokens = len(self.tokens)
1075            self._scan(lambda: self._peek == ";")
1076            self.tokens = self.tokens[:tokens]
1077            text = self.sql[start : self._current].strip()
1078            if text:
1079                self._add(TokenType.STRING, text)
1080
1081    def _scan_keywords(self) -> None:
1082        size = 0
1083        word = None
1084        chars = self._text
1085        char = chars
1086        prev_space = False
1087        skip = False
1088        trie = self._KEYWORD_TRIE
1089        single_token = char in self.SINGLE_TOKENS
1090
1091        while chars:
1092            if skip:
1093                result = TrieResult.PREFIX
1094            else:
1095                result, trie = in_trie(trie, char.upper())
1096
1097            if result == TrieResult.FAILED:
1098                break
1099            if result == TrieResult.EXISTS:
1100                word = chars
1101
1102            end = self._current + size
1103            size += 1
1104
1105            if end < self.size:
1106                char = self.sql[end]
1107                single_token = single_token or char in self.SINGLE_TOKENS
1108                is_space = char.isspace()
1109
1110                if not is_space or not prev_space:
1111                    if is_space:
1112                        char = " "
1113                    chars += char
1114                    prev_space = is_space
1115                    skip = False
1116                else:
1117                    skip = True
1118            else:
1119                char = ""
1120                break
1121
1122        if word:
1123            if self._scan_string(word):
1124                return
1125            if self._scan_comment(word):
1126                return
1127            if prev_space or single_token or not char:
1128                self._advance(size - 1)
1129                word = word.upper()
1130                self._add(self.KEYWORDS[word], text=word)
1131                return
1132
1133        if self._char in self.SINGLE_TOKENS:
1134            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1135            return
1136
1137        self._scan_var()
1138
1139    def _scan_comment(self, comment_start: str) -> bool:
1140        if comment_start not in self._COMMENTS:
1141            return False
1142
1143        comment_start_line = self._line
1144        comment_start_size = len(comment_start)
1145        comment_end = self._COMMENTS[comment_start]
1146
1147        if comment_end:
1148            # Skip the comment's start delimiter
1149            self._advance(comment_start_size)
1150
1151            comment_end_size = len(comment_end)
1152            while not self._end and self._chars(comment_end_size) != comment_end:
1153                self._advance(alnum=True)
1154
1155            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1156            self._advance(comment_end_size - 1)
1157        else:
1158            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1159                self._advance(alnum=True)
1160            self._comments.append(self._text[comment_start_size:])
1161
1162        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1163        # Multiple consecutive comments are preserved by appending them to the current comments list.
1164        if comment_start_line == self._prev_token_line:
1165            self.tokens[-1].comments.extend(self._comments)
1166            self._comments = []
1167            self._prev_token_line = self._line
1168
1169        return True
1170
1171    def _scan_number(self) -> None:
1172        if self._char == "0":
1173            peek = self._peek.upper()
1174            if peek == "B":
1175                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1176            elif peek == "X":
1177                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1178
1179        decimal = False
1180        scientific = 0
1181
1182        while True:
1183            if self._peek.isdigit():
1184                self._advance()
1185            elif self._peek == "." and not decimal:
1186                decimal = True
1187                self._advance()
1188            elif self._peek in ("-", "+") and scientific == 1:
1189                scientific += 1
1190                self._advance()
1191            elif self._peek.upper() == "E" and not scientific:
1192                scientific += 1
1193                self._advance()
1194            elif self._peek.isidentifier():
1195                number_text = self._text
1196                literal = ""
1197
1198                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1199                    literal += self._peek
1200                    self._advance()
1201
1202                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1203
1204                if token_type:
1205                    self._add(TokenType.NUMBER, number_text)
1206                    self._add(TokenType.DCOLON, "::")
1207                    return self._add(token_type, literal)
1208                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1209                    return self._add(TokenType.VAR)
1210
1211                self._advance(-len(literal))
1212                return self._add(TokenType.NUMBER, number_text)
1213            else:
1214                return self._add(TokenType.NUMBER)
1215
1216    def _scan_bits(self) -> None:
1217        self._advance()
1218        value = self._extract_value()
1219        try:
1220            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1221            int(value, 2)
1222            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1223        except ValueError:
1224            self._add(TokenType.IDENTIFIER)
1225
1226    def _scan_hex(self) -> None:
1227        self._advance()
1228        value = self._extract_value()
1229        try:
1230            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1231            int(value, 16)
1232            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1233        except ValueError:
1234            self._add(TokenType.IDENTIFIER)
1235
1236    def _extract_value(self) -> str:
1237        while True:
1238            char = self._peek.strip()
1239            if char and char not in self.SINGLE_TOKENS:
1240                self._advance(alnum=True)
1241            else:
1242                break
1243
1244        return self._text
1245
1246    def _scan_string(self, start: str) -> bool:
1247        base = None
1248        token_type = TokenType.STRING
1249
1250        if start in self._QUOTES:
1251            end = self._QUOTES[start]
1252        elif start in self._FORMAT_STRINGS:
1253            end, token_type = self._FORMAT_STRINGS[start]
1254
1255            if token_type == TokenType.HEX_STRING:
1256                base = 16
1257            elif token_type == TokenType.BIT_STRING:
1258                base = 2
1259            elif token_type == TokenType.HEREDOC_STRING:
1260                if (
1261                    self.HEREDOC_TAG_IS_IDENTIFIER
1262                    and not self._peek.isidentifier()
1263                    and not self._peek == end
1264                ):
1265                    if self.HEREDOC_STRING_ALTERNATIVE != token_type.VAR:
1266                        self._add(self.HEREDOC_STRING_ALTERNATIVE)
1267                    else:
1268                        self._scan_var()
1269
1270                    return True
1271
1272                self._advance()
1273
1274                if self._char == end:
1275                    tag = ""
1276                else:
1277                    tag = self._extract_string(
1278                        end,
1279                        unescape_sequences=False,
1280                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1281                    )
1282
1283                if self._end and tag and self.HEREDOC_TAG_IS_IDENTIFIER:
1284                    self._advance(-len(tag))
1285                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1286                    return True
1287
1288                end = f"{start}{tag}{end}"
1289        else:
1290            return False
1291
1292        self._advance(len(start))
1293        text = self._extract_string(end, unescape_sequences=token_type != TokenType.RAW_STRING)
1294
1295        if base:
1296            try:
1297                int(text, base)
1298            except Exception:
1299                raise TokenError(
1300                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1301                )
1302
1303        self._add(token_type, text)
1304        return True
1305
1306    def _scan_identifier(self, identifier_end: str) -> None:
1307        self._advance()
1308        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1309        self._add(TokenType.IDENTIFIER, text)
1310
1311    def _scan_var(self) -> None:
1312        while True:
1313            char = self._peek.strip()
1314            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1315                self._advance(alnum=True)
1316            else:
1317                break
1318
1319        self._add(
1320            TokenType.VAR
1321            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1322            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1323        )
1324
1325    def _extract_string(
1326        self,
1327        delimiter: str,
1328        escapes: t.Optional[t.Set[str]] = None,
1329        unescape_sequences: bool = True,
1330        raise_unmatched: bool = True,
1331    ) -> str:
1332        text = ""
1333        delim_size = len(delimiter)
1334        escapes = self._STRING_ESCAPES if escapes is None else escapes
1335
1336        while True:
1337            if (
1338                unescape_sequences
1339                and self.dialect.UNESCAPED_SEQUENCES
1340                and self._peek
1341                and self._char in self.STRING_ESCAPES
1342            ):
1343                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1344                if unescaped_sequence:
1345                    self._advance(2)
1346                    text += unescaped_sequence
1347                    continue
1348            if (
1349                self._char in escapes
1350                and (self._peek == delimiter or self._peek in escapes)
1351                and (self._char not in self._QUOTES or self._char == self._peek)
1352            ):
1353                if self._peek == delimiter:
1354                    text += self._peek
1355                else:
1356                    text += self._char + self._peek
1357
1358                if self._current + 1 < self.size:
1359                    self._advance(2)
1360                else:
1361                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1362            else:
1363                if self._chars(delim_size) == delimiter:
1364                    if delim_size > 1:
1365                        self._advance(delim_size - 1)
1366                    break
1367
1368                if self._end:
1369                    if not raise_unmatched:
1370                        return text + self._char
1371
1372                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1373
1374                current = self._current - 1
1375                self._advance(alnum=True)
1376                text += self.sql[current : self._current - 1]
1377
1378        return text
1379
1380    def tokenize_rs(self, sql: str) -> t.List[Token]:
1381        if not self._RS_TOKENIZER:
1382            raise SqlglotError("Rust tokenizer is not available")
1383
1384        try:
1385            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1386            for token in tokens:
1387                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1388            return tokens
1389        except Exception as e:
1390            raise TokenError(str(e))
Tokenizer( dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None)
921    def __init__(self, dialect: DialectType = None) -> None:
922        from sqlglot.dialects import Dialect
923
924        self.dialect = Dialect.get_or_raise(dialect)
925
926        if USE_RS_TOKENIZER:
927            self._rs_dialect_settings = RsTokenizerDialectSettings(
928                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
929                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
930            )
931
932        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.HASH: 'HASH'>, "'": <TokenType.UNKNOWN: 'UNKNOWN'>, '`': <TokenType.UNKNOWN: 'UNKNOWN'>, '"': <TokenType.UNKNOWN: 'UNKNOWN'>}
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'>, 'COPY': <TokenType.COPY: 'COPY'>, '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'>, 'UINT': <TokenType.UINT: 'UINT'>, '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'>, '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.FETCH: 'FETCH'>, <TokenType.SHOW: 'SHOW'>, <TokenType.COMMAND: 'COMMAND'>, <TokenType.EXECUTE: 'EXECUTE'>}
COMMAND_PREFIX_TOKENS = {<TokenType.BEGIN: 'BEGIN'>, <TokenType.SEMICOLON: 'SEMICOLON'>}
NUMERIC_LITERALS: Dict[str, str] = {}
COMMENTS = ['--', ('/*', '*/')]
dialect
def reset(self) -> None:
934    def reset(self) -> None:
935        self.sql = ""
936        self.size = 0
937        self.tokens: t.List[Token] = []
938        self._start = 0
939        self._current = 0
940        self._line = 1
941        self._col = 0
942        self._comments: t.List[str] = []
943
944        self._char = ""
945        self._end = False
946        self._peek = ""
947        self._prev_token_line = -1
def tokenize(self, sql: str) -> List[Token]:
949    def tokenize(self, sql: str) -> t.List[Token]:
950        """Returns a list of tokens corresponding to the SQL string `sql`."""
951        if USE_RS_TOKENIZER:
952            return self.tokenize_rs(sql)
953
954        self.reset()
955        self.sql = sql
956        self.size = len(sql)
957
958        try:
959            self._scan()
960        except Exception as e:
961            start = max(self._current - 50, 0)
962            end = min(self._current + 50, self.size - 1)
963            context = self.sql[start:end]
964            raise TokenError(f"Error tokenizing '{context}'") from e
965
966        return self.tokens

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

def tokenize_rs(self, sql: str) -> List[Token]:
1380    def tokenize_rs(self, sql: str) -> t.List[Token]:
1381        if not self._RS_TOKENIZER:
1382            raise SqlglotError("Rust tokenizer is not available")
1383
1384        try:
1385            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1386            for token in tokens:
1387                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1388            return tokens
1389        except Exception as e:
1390            raise TokenError(str(e))
size
sql
tokens