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    WAREHOUSE = auto()
  96    STREAMLIT = auto()
  97    VAR = auto()
  98    BIT_STRING = auto()
  99    HEX_STRING = auto()
 100    BYTE_STRING = auto()
 101    NATIONAL_STRING = auto()
 102    RAW_STRING = auto()
 103    HEREDOC_STRING = auto()
 104    UNICODE_STRING = auto()
 105
 106    # types
 107    BIT = auto()
 108    BOOLEAN = auto()
 109    TINYINT = auto()
 110    UTINYINT = auto()
 111    SMALLINT = auto()
 112    USMALLINT = auto()
 113    MEDIUMINT = auto()
 114    UMEDIUMINT = auto()
 115    INT = auto()
 116    UINT = auto()
 117    BIGINT = auto()
 118    UBIGINT = auto()
 119    INT128 = auto()
 120    UINT128 = auto()
 121    INT256 = auto()
 122    UINT256 = auto()
 123    FLOAT = auto()
 124    DOUBLE = auto()
 125    DECIMAL = auto()
 126    DECIMAL32 = auto()
 127    DECIMAL64 = auto()
 128    DECIMAL128 = auto()
 129    UDECIMAL = auto()
 130    BIGDECIMAL = auto()
 131    CHAR = auto()
 132    NCHAR = auto()
 133    VARCHAR = auto()
 134    NVARCHAR = auto()
 135    BPCHAR = auto()
 136    TEXT = auto()
 137    MEDIUMTEXT = auto()
 138    LONGTEXT = auto()
 139    MEDIUMBLOB = auto()
 140    LONGBLOB = auto()
 141    TINYBLOB = auto()
 142    TINYTEXT = auto()
 143    NAME = auto()
 144    BINARY = auto()
 145    VARBINARY = auto()
 146    JSON = auto()
 147    JSONB = auto()
 148    TIME = auto()
 149    TIMETZ = auto()
 150    TIMESTAMP = auto()
 151    TIMESTAMPTZ = auto()
 152    TIMESTAMPLTZ = auto()
 153    TIMESTAMPNTZ = auto()
 154    TIMESTAMP_S = auto()
 155    TIMESTAMP_MS = auto()
 156    TIMESTAMP_NS = auto()
 157    DATETIME = auto()
 158    DATETIME64 = auto()
 159    DATE = auto()
 160    DATE32 = auto()
 161    INT4RANGE = auto()
 162    INT4MULTIRANGE = auto()
 163    INT8RANGE = auto()
 164    INT8MULTIRANGE = auto()
 165    NUMRANGE = auto()
 166    NUMMULTIRANGE = auto()
 167    TSRANGE = auto()
 168    TSMULTIRANGE = auto()
 169    TSTZRANGE = auto()
 170    TSTZMULTIRANGE = auto()
 171    DATERANGE = auto()
 172    DATEMULTIRANGE = auto()
 173    UUID = auto()
 174    GEOGRAPHY = auto()
 175    NULLABLE = auto()
 176    GEOMETRY = auto()
 177    HLLSKETCH = auto()
 178    HSTORE = auto()
 179    SUPER = auto()
 180    SERIAL = auto()
 181    SMALLSERIAL = auto()
 182    BIGSERIAL = auto()
 183    XML = auto()
 184    YEAR = auto()
 185    UNIQUEIDENTIFIER = auto()
 186    USERDEFINED = auto()
 187    MONEY = auto()
 188    SMALLMONEY = auto()
 189    ROWVERSION = auto()
 190    IMAGE = auto()
 191    VARIANT = auto()
 192    OBJECT = auto()
 193    INET = auto()
 194    IPADDRESS = auto()
 195    IPPREFIX = auto()
 196    IPV4 = auto()
 197    IPV6 = auto()
 198    ENUM = auto()
 199    ENUM8 = auto()
 200    ENUM16 = auto()
 201    FIXEDSTRING = auto()
 202    LOWCARDINALITY = auto()
 203    NESTED = auto()
 204    AGGREGATEFUNCTION = auto()
 205    SIMPLEAGGREGATEFUNCTION = auto()
 206    TDIGEST = auto()
 207    UNKNOWN = auto()
 208    VECTOR = auto()
 209
 210    # keywords
 211    ALIAS = auto()
 212    ALTER = auto()
 213    ALWAYS = auto()
 214    ALL = auto()
 215    ANTI = auto()
 216    ANY = auto()
 217    APPLY = auto()
 218    ARRAY = auto()
 219    ASC = auto()
 220    ASOF = auto()
 221    AUTO_INCREMENT = auto()
 222    BEGIN = auto()
 223    BETWEEN = auto()
 224    BULK_COLLECT_INTO = auto()
 225    CACHE = auto()
 226    CASE = auto()
 227    CHARACTER_SET = auto()
 228    CLUSTER_BY = auto()
 229    COLLATE = auto()
 230    COMMAND = auto()
 231    COMMENT = auto()
 232    COMMIT = auto()
 233    CONNECT_BY = auto()
 234    CONSTRAINT = auto()
 235    COPY = auto()
 236    CREATE = auto()
 237    CROSS = auto()
 238    CUBE = auto()
 239    CURRENT_DATE = auto()
 240    CURRENT_DATETIME = auto()
 241    CURRENT_TIME = auto()
 242    CURRENT_TIMESTAMP = auto()
 243    CURRENT_USER = auto()
 244    DECLARE = auto()
 245    DEFAULT = auto()
 246    DELETE = auto()
 247    DESC = auto()
 248    DESCRIBE = auto()
 249    DICTIONARY = auto()
 250    DISTINCT = auto()
 251    DISTRIBUTE_BY = auto()
 252    DIV = auto()
 253    DROP = auto()
 254    ELSE = auto()
 255    END = auto()
 256    ESCAPE = auto()
 257    EXCEPT = auto()
 258    EXECUTE = auto()
 259    EXISTS = auto()
 260    FALSE = auto()
 261    FETCH = auto()
 262    FILTER = auto()
 263    FINAL = auto()
 264    FIRST = auto()
 265    FOR = auto()
 266    FORCE = auto()
 267    FOREIGN_KEY = auto()
 268    FORMAT = auto()
 269    FROM = auto()
 270    FULL = auto()
 271    FUNCTION = auto()
 272    GLOB = auto()
 273    GLOBAL = auto()
 274    GRANT = auto()
 275    GROUP_BY = auto()
 276    GROUPING_SETS = auto()
 277    HAVING = auto()
 278    HINT = auto()
 279    IGNORE = auto()
 280    ILIKE = auto()
 281    ILIKE_ANY = auto()
 282    IN = auto()
 283    INDEX = auto()
 284    INNER = auto()
 285    INSERT = auto()
 286    INTERSECT = auto()
 287    INTERVAL = auto()
 288    INTO = auto()
 289    INTRODUCER = auto()
 290    IRLIKE = auto()
 291    IS = auto()
 292    ISNULL = auto()
 293    JOIN = auto()
 294    JOIN_MARKER = auto()
 295    KEEP = auto()
 296    KEY = auto()
 297    KILL = auto()
 298    LANGUAGE = auto()
 299    LATERAL = auto()
 300    LEFT = auto()
 301    LIKE = auto()
 302    LIKE_ANY = auto()
 303    LIMIT = auto()
 304    LIST = auto()
 305    LOAD = auto()
 306    LOCK = auto()
 307    MAP = auto()
 308    MATCH_CONDITION = auto()
 309    MATCH_RECOGNIZE = auto()
 310    MEMBER_OF = auto()
 311    MERGE = auto()
 312    MOD = auto()
 313    MODEL = auto()
 314    NATURAL = auto()
 315    NEXT = auto()
 316    NOTNULL = auto()
 317    NULL = auto()
 318    OBJECT_IDENTIFIER = auto()
 319    OFFSET = auto()
 320    ON = auto()
 321    ONLY = auto()
 322    OPERATOR = auto()
 323    ORDER_BY = auto()
 324    ORDER_SIBLINGS_BY = auto()
 325    ORDERED = auto()
 326    ORDINALITY = auto()
 327    OUTER = auto()
 328    OVER = auto()
 329    OVERLAPS = auto()
 330    OVERWRITE = auto()
 331    PARTITION = auto()
 332    PARTITION_BY = auto()
 333    PERCENT = auto()
 334    PIVOT = auto()
 335    PLACEHOLDER = auto()
 336    POSITIONAL = auto()
 337    PRAGMA = auto()
 338    PREWHERE = auto()
 339    PRIMARY_KEY = auto()
 340    PROCEDURE = auto()
 341    PROPERTIES = auto()
 342    PSEUDO_TYPE = auto()
 343    QUALIFY = auto()
 344    QUOTE = auto()
 345    RANGE = auto()
 346    RECURSIVE = auto()
 347    REFRESH = auto()
 348    RENAME = auto()
 349    REPLACE = auto()
 350    RETURNING = auto()
 351    REFERENCES = auto()
 352    RIGHT = auto()
 353    RLIKE = auto()
 354    ROLLBACK = auto()
 355    ROLLUP = auto()
 356    ROW = auto()
 357    ROWS = auto()
 358    SELECT = auto()
 359    SEMI = auto()
 360    SEPARATOR = auto()
 361    SEQUENCE = auto()
 362    SERDE_PROPERTIES = auto()
 363    SET = auto()
 364    SETTINGS = auto()
 365    SHOW = auto()
 366    SIMILAR_TO = auto()
 367    SOME = auto()
 368    SORT_BY = auto()
 369    START_WITH = auto()
 370    STORAGE_INTEGRATION = auto()
 371    STRAIGHT_JOIN = auto()
 372    STRUCT = auto()
 373    SUMMARIZE = auto()
 374    TABLE_SAMPLE = auto()
 375    TAG = auto()
 376    TEMPORARY = auto()
 377    TOP = auto()
 378    THEN = auto()
 379    TRUE = auto()
 380    TRUNCATE = auto()
 381    UNCACHE = auto()
 382    UNION = auto()
 383    UNNEST = auto()
 384    UNPIVOT = auto()
 385    UPDATE = auto()
 386    USE = auto()
 387    USING = auto()
 388    VALUES = auto()
 389    VIEW = auto()
 390    VOLATILE = auto()
 391    WHEN = auto()
 392    WHERE = auto()
 393    WINDOW = auto()
 394    WITH = auto()
 395    UNIQUE = auto()
 396    VERSION_SNAPSHOT = auto()
 397    TIMESTAMP_SNAPSHOT = auto()
 398    OPTION = auto()
 399
 400
 401_ALL_TOKEN_TYPES = list(TokenType)
 402_TOKEN_TYPE_TO_INDEX = {token_type: i for i, token_type in enumerate(_ALL_TOKEN_TYPES)}
 403
 404
 405class Token:
 406    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
 407
 408    @classmethod
 409    def number(cls, number: int) -> Token:
 410        """Returns a NUMBER token with `number` as its text."""
 411        return cls(TokenType.NUMBER, str(number))
 412
 413    @classmethod
 414    def string(cls, string: str) -> Token:
 415        """Returns a STRING token with `string` as its text."""
 416        return cls(TokenType.STRING, string)
 417
 418    @classmethod
 419    def identifier(cls, identifier: str) -> Token:
 420        """Returns an IDENTIFIER token with `identifier` as its text."""
 421        return cls(TokenType.IDENTIFIER, identifier)
 422
 423    @classmethod
 424    def var(cls, var: str) -> Token:
 425        """Returns an VAR token with `var` as its text."""
 426        return cls(TokenType.VAR, var)
 427
 428    def __init__(
 429        self,
 430        token_type: TokenType,
 431        text: str,
 432        line: int = 1,
 433        col: int = 1,
 434        start: int = 0,
 435        end: int = 0,
 436        comments: t.Optional[t.List[str]] = None,
 437    ) -> None:
 438        """Token initializer.
 439
 440        Args:
 441            token_type: The TokenType Enum.
 442            text: The text of the token.
 443            line: The line that the token ends on.
 444            col: The column that the token ends on.
 445            start: The start index of the token.
 446            end: The ending index of the token.
 447            comments: The comments to attach to the token.
 448        """
 449        self.token_type = token_type
 450        self.text = text
 451        self.line = line
 452        self.col = col
 453        self.start = start
 454        self.end = end
 455        self.comments = [] if comments is None else comments
 456
 457    def __repr__(self) -> str:
 458        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
 459        return f"<Token {attributes}>"
 460
 461
 462class _Tokenizer(type):
 463    def __new__(cls, clsname, bases, attrs):
 464        klass = super().__new__(cls, clsname, bases, attrs)
 465
 466        def _convert_quotes(arr: t.List[str | t.Tuple[str, str]]) -> t.Dict[str, str]:
 467            return dict(
 468                (item, item) if isinstance(item, str) else (item[0], item[1]) for item in arr
 469            )
 470
 471        def _quotes_to_format(
 472            token_type: TokenType, arr: t.List[str | t.Tuple[str, str]]
 473        ) -> t.Dict[str, t.Tuple[str, TokenType]]:
 474            return {k: (v, token_type) for k, v in _convert_quotes(arr).items()}
 475
 476        klass._QUOTES = _convert_quotes(klass.QUOTES)
 477        klass._IDENTIFIERS = _convert_quotes(klass.IDENTIFIERS)
 478
 479        klass._FORMAT_STRINGS = {
 480            **{
 481                p + s: (e, TokenType.NATIONAL_STRING)
 482                for s, e in klass._QUOTES.items()
 483                for p in ("n", "N")
 484            },
 485            **_quotes_to_format(TokenType.BIT_STRING, klass.BIT_STRINGS),
 486            **_quotes_to_format(TokenType.BYTE_STRING, klass.BYTE_STRINGS),
 487            **_quotes_to_format(TokenType.HEX_STRING, klass.HEX_STRINGS),
 488            **_quotes_to_format(TokenType.RAW_STRING, klass.RAW_STRINGS),
 489            **_quotes_to_format(TokenType.HEREDOC_STRING, klass.HEREDOC_STRINGS),
 490            **_quotes_to_format(TokenType.UNICODE_STRING, klass.UNICODE_STRINGS),
 491        }
 492
 493        klass._STRING_ESCAPES = set(klass.STRING_ESCAPES)
 494        klass._IDENTIFIER_ESCAPES = set(klass.IDENTIFIER_ESCAPES)
 495        klass._COMMENTS = {
 496            **dict(
 497                (comment, None) if isinstance(comment, str) else (comment[0], comment[1])
 498                for comment in klass.COMMENTS
 499            ),
 500            "{#": "#}",  # Ensure Jinja comments are tokenized correctly in all dialects
 501        }
 502
 503        klass._KEYWORD_TRIE = new_trie(
 504            key.upper()
 505            for key in (
 506                *klass.KEYWORDS,
 507                *klass._COMMENTS,
 508                *klass._QUOTES,
 509                *klass._FORMAT_STRINGS,
 510            )
 511            if " " in key or any(single in key for single in klass.SINGLE_TOKENS)
 512        )
 513
 514        if USE_RS_TOKENIZER:
 515            settings = RsTokenizerSettings(
 516                white_space={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.WHITE_SPACE.items()},
 517                single_tokens={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.SINGLE_TOKENS.items()},
 518                keywords={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.KEYWORDS.items()},
 519                numeric_literals=klass.NUMERIC_LITERALS,
 520                identifiers=klass._IDENTIFIERS,
 521                identifier_escapes=klass._IDENTIFIER_ESCAPES,
 522                string_escapes=klass._STRING_ESCAPES,
 523                quotes=klass._QUOTES,
 524                format_strings={
 525                    k: (v1, _TOKEN_TYPE_TO_INDEX[v2])
 526                    for k, (v1, v2) in klass._FORMAT_STRINGS.items()
 527                },
 528                has_bit_strings=bool(klass.BIT_STRINGS),
 529                has_hex_strings=bool(klass.HEX_STRINGS),
 530                comments=klass._COMMENTS,
 531                var_single_tokens=klass.VAR_SINGLE_TOKENS,
 532                commands={_TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMANDS},
 533                command_prefix_tokens={
 534                    _TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMAND_PREFIX_TOKENS
 535                },
 536                heredoc_tag_is_identifier=klass.HEREDOC_TAG_IS_IDENTIFIER,
 537                string_escapes_allowed_in_raw_strings=klass.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS,
 538                nested_comments=klass.NESTED_COMMENTS,
 539            )
 540            token_types = RsTokenTypeSettings(
 541                bit_string=_TOKEN_TYPE_TO_INDEX[TokenType.BIT_STRING],
 542                break_=_TOKEN_TYPE_TO_INDEX[TokenType.BREAK],
 543                dcolon=_TOKEN_TYPE_TO_INDEX[TokenType.DCOLON],
 544                heredoc_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEREDOC_STRING],
 545                raw_string=_TOKEN_TYPE_TO_INDEX[TokenType.RAW_STRING],
 546                hex_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEX_STRING],
 547                identifier=_TOKEN_TYPE_TO_INDEX[TokenType.IDENTIFIER],
 548                number=_TOKEN_TYPE_TO_INDEX[TokenType.NUMBER],
 549                parameter=_TOKEN_TYPE_TO_INDEX[TokenType.PARAMETER],
 550                semicolon=_TOKEN_TYPE_TO_INDEX[TokenType.SEMICOLON],
 551                string=_TOKEN_TYPE_TO_INDEX[TokenType.STRING],
 552                var=_TOKEN_TYPE_TO_INDEX[TokenType.VAR],
 553                heredoc_string_alternative=_TOKEN_TYPE_TO_INDEX[klass.HEREDOC_STRING_ALTERNATIVE],
 554            )
 555            klass._RS_TOKENIZER = RsTokenizer(settings, token_types)
 556        else:
 557            klass._RS_TOKENIZER = None
 558
 559        return klass
 560
 561
 562class Tokenizer(metaclass=_Tokenizer):
 563    SINGLE_TOKENS = {
 564        "(": TokenType.L_PAREN,
 565        ")": TokenType.R_PAREN,
 566        "[": TokenType.L_BRACKET,
 567        "]": TokenType.R_BRACKET,
 568        "{": TokenType.L_BRACE,
 569        "}": TokenType.R_BRACE,
 570        "&": TokenType.AMP,
 571        "^": TokenType.CARET,
 572        ":": TokenType.COLON,
 573        ",": TokenType.COMMA,
 574        ".": TokenType.DOT,
 575        "-": TokenType.DASH,
 576        "=": TokenType.EQ,
 577        ">": TokenType.GT,
 578        "<": TokenType.LT,
 579        "%": TokenType.MOD,
 580        "!": TokenType.NOT,
 581        "|": TokenType.PIPE,
 582        "+": TokenType.PLUS,
 583        ";": TokenType.SEMICOLON,
 584        "/": TokenType.SLASH,
 585        "\\": TokenType.BACKSLASH,
 586        "*": TokenType.STAR,
 587        "~": TokenType.TILDA,
 588        "?": TokenType.PLACEHOLDER,
 589        "@": TokenType.PARAMETER,
 590        "#": TokenType.HASH,
 591        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 592        "'": TokenType.UNKNOWN,
 593        "`": TokenType.UNKNOWN,
 594        '"': TokenType.UNKNOWN,
 595    }
 596
 597    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 598    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 599    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 600    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 601    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 602    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 603    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 604    IDENTIFIER_ESCAPES = ['"']
 605    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 606    STRING_ESCAPES = ["'"]
 607    VAR_SINGLE_TOKENS: t.Set[str] = set()
 608
 609    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 610    HEREDOC_TAG_IS_IDENTIFIER = False
 611
 612    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 613    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 614
 615    # Whether string escape characters function as such when placed within raw strings
 616    STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS = True
 617
 618    NESTED_COMMENTS = True
 619
 620    # Autofilled
 621    _COMMENTS: t.Dict[str, str] = {}
 622    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 623    _IDENTIFIERS: t.Dict[str, str] = {}
 624    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 625    _QUOTES: t.Dict[str, str] = {}
 626    _STRING_ESCAPES: t.Set[str] = set()
 627    _KEYWORD_TRIE: t.Dict = {}
 628    _RS_TOKENIZER: t.Optional[t.Any] = None
 629
 630    KEYWORDS: t.Dict[str, TokenType] = {
 631        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 632        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 633        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 634        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 635        "/*+": TokenType.HINT,
 636        "==": TokenType.EQ,
 637        "::": TokenType.DCOLON,
 638        "||": TokenType.DPIPE,
 639        ">=": TokenType.GTE,
 640        "<=": TokenType.LTE,
 641        "<>": TokenType.NEQ,
 642        "!=": TokenType.NEQ,
 643        ":=": TokenType.COLON_EQ,
 644        "<=>": TokenType.NULLSAFE_EQ,
 645        "->": TokenType.ARROW,
 646        "->>": TokenType.DARROW,
 647        "=>": TokenType.FARROW,
 648        "#>": TokenType.HASH_ARROW,
 649        "#>>": TokenType.DHASH_ARROW,
 650        "<->": TokenType.LR_ARROW,
 651        "&&": TokenType.DAMP,
 652        "??": TokenType.DQMARK,
 653        "~~~": TokenType.GLOB,
 654        "~~": TokenType.LIKE,
 655        "ALL": TokenType.ALL,
 656        "ALWAYS": TokenType.ALWAYS,
 657        "AND": TokenType.AND,
 658        "ANTI": TokenType.ANTI,
 659        "ANY": TokenType.ANY,
 660        "ASC": TokenType.ASC,
 661        "AS": TokenType.ALIAS,
 662        "ASOF": TokenType.ASOF,
 663        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 664        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 665        "BEGIN": TokenType.BEGIN,
 666        "BETWEEN": TokenType.BETWEEN,
 667        "CACHE": TokenType.CACHE,
 668        "UNCACHE": TokenType.UNCACHE,
 669        "CASE": TokenType.CASE,
 670        "CHARACTER SET": TokenType.CHARACTER_SET,
 671        "CLUSTER BY": TokenType.CLUSTER_BY,
 672        "COLLATE": TokenType.COLLATE,
 673        "COLUMN": TokenType.COLUMN,
 674        "COMMIT": TokenType.COMMIT,
 675        "CONNECT BY": TokenType.CONNECT_BY,
 676        "CONSTRAINT": TokenType.CONSTRAINT,
 677        "COPY": TokenType.COPY,
 678        "CREATE": TokenType.CREATE,
 679        "CROSS": TokenType.CROSS,
 680        "CUBE": TokenType.CUBE,
 681        "CURRENT_DATE": TokenType.CURRENT_DATE,
 682        "CURRENT_TIME": TokenType.CURRENT_TIME,
 683        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 684        "CURRENT_USER": TokenType.CURRENT_USER,
 685        "DATABASE": TokenType.DATABASE,
 686        "DEFAULT": TokenType.DEFAULT,
 687        "DELETE": TokenType.DELETE,
 688        "DESC": TokenType.DESC,
 689        "DESCRIBE": TokenType.DESCRIBE,
 690        "DISTINCT": TokenType.DISTINCT,
 691        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 692        "DIV": TokenType.DIV,
 693        "DROP": TokenType.DROP,
 694        "ELSE": TokenType.ELSE,
 695        "END": TokenType.END,
 696        "ENUM": TokenType.ENUM,
 697        "ESCAPE": TokenType.ESCAPE,
 698        "EXCEPT": TokenType.EXCEPT,
 699        "EXECUTE": TokenType.EXECUTE,
 700        "EXISTS": TokenType.EXISTS,
 701        "FALSE": TokenType.FALSE,
 702        "FETCH": TokenType.FETCH,
 703        "FILTER": TokenType.FILTER,
 704        "FIRST": TokenType.FIRST,
 705        "FULL": TokenType.FULL,
 706        "FUNCTION": TokenType.FUNCTION,
 707        "FOR": TokenType.FOR,
 708        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 709        "FORMAT": TokenType.FORMAT,
 710        "FROM": TokenType.FROM,
 711        "GEOGRAPHY": TokenType.GEOGRAPHY,
 712        "GEOMETRY": TokenType.GEOMETRY,
 713        "GLOB": TokenType.GLOB,
 714        "GROUP BY": TokenType.GROUP_BY,
 715        "GROUPING SETS": TokenType.GROUPING_SETS,
 716        "HAVING": TokenType.HAVING,
 717        "ILIKE": TokenType.ILIKE,
 718        "IN": TokenType.IN,
 719        "INDEX": TokenType.INDEX,
 720        "INET": TokenType.INET,
 721        "INNER": TokenType.INNER,
 722        "INSERT": TokenType.INSERT,
 723        "INTERVAL": TokenType.INTERVAL,
 724        "INTERSECT": TokenType.INTERSECT,
 725        "INTO": TokenType.INTO,
 726        "IS": TokenType.IS,
 727        "ISNULL": TokenType.ISNULL,
 728        "JOIN": TokenType.JOIN,
 729        "KEEP": TokenType.KEEP,
 730        "KILL": TokenType.KILL,
 731        "LATERAL": TokenType.LATERAL,
 732        "LEFT": TokenType.LEFT,
 733        "LIKE": TokenType.LIKE,
 734        "LIMIT": TokenType.LIMIT,
 735        "LOAD": TokenType.LOAD,
 736        "LOCK": TokenType.LOCK,
 737        "MERGE": TokenType.MERGE,
 738        "NATURAL": TokenType.NATURAL,
 739        "NEXT": TokenType.NEXT,
 740        "NOT": TokenType.NOT,
 741        "NOTNULL": TokenType.NOTNULL,
 742        "NULL": TokenType.NULL,
 743        "OBJECT": TokenType.OBJECT,
 744        "OFFSET": TokenType.OFFSET,
 745        "ON": TokenType.ON,
 746        "OR": TokenType.OR,
 747        "XOR": TokenType.XOR,
 748        "ORDER BY": TokenType.ORDER_BY,
 749        "ORDINALITY": TokenType.ORDINALITY,
 750        "OUTER": TokenType.OUTER,
 751        "OVER": TokenType.OVER,
 752        "OVERLAPS": TokenType.OVERLAPS,
 753        "OVERWRITE": TokenType.OVERWRITE,
 754        "PARTITION": TokenType.PARTITION,
 755        "PARTITION BY": TokenType.PARTITION_BY,
 756        "PARTITIONED BY": TokenType.PARTITION_BY,
 757        "PARTITIONED_BY": TokenType.PARTITION_BY,
 758        "PERCENT": TokenType.PERCENT,
 759        "PIVOT": TokenType.PIVOT,
 760        "PRAGMA": TokenType.PRAGMA,
 761        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 762        "PROCEDURE": TokenType.PROCEDURE,
 763        "QUALIFY": TokenType.QUALIFY,
 764        "RANGE": TokenType.RANGE,
 765        "RECURSIVE": TokenType.RECURSIVE,
 766        "REGEXP": TokenType.RLIKE,
 767        "RENAME": TokenType.RENAME,
 768        "REPLACE": TokenType.REPLACE,
 769        "RETURNING": TokenType.RETURNING,
 770        "REFERENCES": TokenType.REFERENCES,
 771        "RIGHT": TokenType.RIGHT,
 772        "RLIKE": TokenType.RLIKE,
 773        "ROLLBACK": TokenType.ROLLBACK,
 774        "ROLLUP": TokenType.ROLLUP,
 775        "ROW": TokenType.ROW,
 776        "ROWS": TokenType.ROWS,
 777        "SCHEMA": TokenType.SCHEMA,
 778        "SELECT": TokenType.SELECT,
 779        "SEMI": TokenType.SEMI,
 780        "SET": TokenType.SET,
 781        "SETTINGS": TokenType.SETTINGS,
 782        "SHOW": TokenType.SHOW,
 783        "SIMILAR TO": TokenType.SIMILAR_TO,
 784        "SOME": TokenType.SOME,
 785        "SORT BY": TokenType.SORT_BY,
 786        "START WITH": TokenType.START_WITH,
 787        "STRAIGHT_JOIN": TokenType.STRAIGHT_JOIN,
 788        "TABLE": TokenType.TABLE,
 789        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 790        "TEMP": TokenType.TEMPORARY,
 791        "TEMPORARY": TokenType.TEMPORARY,
 792        "THEN": TokenType.THEN,
 793        "TRUE": TokenType.TRUE,
 794        "TRUNCATE": TokenType.TRUNCATE,
 795        "UNION": TokenType.UNION,
 796        "UNKNOWN": TokenType.UNKNOWN,
 797        "UNNEST": TokenType.UNNEST,
 798        "UNPIVOT": TokenType.UNPIVOT,
 799        "UPDATE": TokenType.UPDATE,
 800        "USE": TokenType.USE,
 801        "USING": TokenType.USING,
 802        "UUID": TokenType.UUID,
 803        "VALUES": TokenType.VALUES,
 804        "VIEW": TokenType.VIEW,
 805        "VOLATILE": TokenType.VOLATILE,
 806        "WHEN": TokenType.WHEN,
 807        "WHERE": TokenType.WHERE,
 808        "WINDOW": TokenType.WINDOW,
 809        "WITH": TokenType.WITH,
 810        "APPLY": TokenType.APPLY,
 811        "ARRAY": TokenType.ARRAY,
 812        "BIT": TokenType.BIT,
 813        "BOOL": TokenType.BOOLEAN,
 814        "BOOLEAN": TokenType.BOOLEAN,
 815        "BYTE": TokenType.TINYINT,
 816        "MEDIUMINT": TokenType.MEDIUMINT,
 817        "INT1": TokenType.TINYINT,
 818        "TINYINT": TokenType.TINYINT,
 819        "INT16": TokenType.SMALLINT,
 820        "SHORT": TokenType.SMALLINT,
 821        "SMALLINT": TokenType.SMALLINT,
 822        "INT128": TokenType.INT128,
 823        "HUGEINT": TokenType.INT128,
 824        "UHUGEINT": TokenType.UINT128,
 825        "INT2": TokenType.SMALLINT,
 826        "INTEGER": TokenType.INT,
 827        "INT": TokenType.INT,
 828        "INT4": TokenType.INT,
 829        "INT32": TokenType.INT,
 830        "INT64": TokenType.BIGINT,
 831        "LONG": TokenType.BIGINT,
 832        "BIGINT": TokenType.BIGINT,
 833        "INT8": TokenType.TINYINT,
 834        "UINT": TokenType.UINT,
 835        "DEC": TokenType.DECIMAL,
 836        "DECIMAL": TokenType.DECIMAL,
 837        "DECIMAL32": TokenType.DECIMAL32,
 838        "DECIMAL64": TokenType.DECIMAL64,
 839        "DECIMAL128": TokenType.DECIMAL128,
 840        "BIGDECIMAL": TokenType.BIGDECIMAL,
 841        "BIGNUMERIC": TokenType.BIGDECIMAL,
 842        "LIST": TokenType.LIST,
 843        "MAP": TokenType.MAP,
 844        "NULLABLE": TokenType.NULLABLE,
 845        "NUMBER": TokenType.DECIMAL,
 846        "NUMERIC": TokenType.DECIMAL,
 847        "FIXED": TokenType.DECIMAL,
 848        "REAL": TokenType.FLOAT,
 849        "FLOAT": TokenType.FLOAT,
 850        "FLOAT4": TokenType.FLOAT,
 851        "FLOAT8": TokenType.DOUBLE,
 852        "DOUBLE": TokenType.DOUBLE,
 853        "DOUBLE PRECISION": TokenType.DOUBLE,
 854        "JSON": TokenType.JSON,
 855        "JSONB": TokenType.JSONB,
 856        "CHAR": TokenType.CHAR,
 857        "CHARACTER": TokenType.CHAR,
 858        "NCHAR": TokenType.NCHAR,
 859        "VARCHAR": TokenType.VARCHAR,
 860        "VARCHAR2": TokenType.VARCHAR,
 861        "NVARCHAR": TokenType.NVARCHAR,
 862        "NVARCHAR2": TokenType.NVARCHAR,
 863        "BPCHAR": TokenType.BPCHAR,
 864        "STR": TokenType.TEXT,
 865        "STRING": TokenType.TEXT,
 866        "TEXT": TokenType.TEXT,
 867        "LONGTEXT": TokenType.LONGTEXT,
 868        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 869        "TINYTEXT": TokenType.TINYTEXT,
 870        "CLOB": TokenType.TEXT,
 871        "LONGVARCHAR": TokenType.TEXT,
 872        "BINARY": TokenType.BINARY,
 873        "BLOB": TokenType.VARBINARY,
 874        "LONGBLOB": TokenType.LONGBLOB,
 875        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 876        "TINYBLOB": TokenType.TINYBLOB,
 877        "BYTEA": TokenType.VARBINARY,
 878        "VARBINARY": TokenType.VARBINARY,
 879        "TIME": TokenType.TIME,
 880        "TIMETZ": TokenType.TIMETZ,
 881        "TIMESTAMP": TokenType.TIMESTAMP,
 882        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 883        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 884        "TIMESTAMP_LTZ": TokenType.TIMESTAMPLTZ,
 885        "TIMESTAMPNTZ": TokenType.TIMESTAMPNTZ,
 886        "TIMESTAMP_NTZ": TokenType.TIMESTAMPNTZ,
 887        "DATE": TokenType.DATE,
 888        "DATETIME": TokenType.DATETIME,
 889        "INT4RANGE": TokenType.INT4RANGE,
 890        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 891        "INT8RANGE": TokenType.INT8RANGE,
 892        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 893        "NUMRANGE": TokenType.NUMRANGE,
 894        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 895        "TSRANGE": TokenType.TSRANGE,
 896        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 897        "TSTZRANGE": TokenType.TSTZRANGE,
 898        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 899        "DATERANGE": TokenType.DATERANGE,
 900        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 901        "UNIQUE": TokenType.UNIQUE,
 902        "VECTOR": TokenType.VECTOR,
 903        "STRUCT": TokenType.STRUCT,
 904        "SEQUENCE": TokenType.SEQUENCE,
 905        "VARIANT": TokenType.VARIANT,
 906        "ALTER": TokenType.ALTER,
 907        "ANALYZE": TokenType.COMMAND,
 908        "CALL": TokenType.COMMAND,
 909        "COMMENT": TokenType.COMMENT,
 910        "EXPLAIN": TokenType.COMMAND,
 911        "GRANT": TokenType.GRANT,
 912        "OPTIMIZE": TokenType.COMMAND,
 913        "PREPARE": TokenType.COMMAND,
 914        "VACUUM": TokenType.COMMAND,
 915        "USER-DEFINED": TokenType.USERDEFINED,
 916        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 917        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 918    }
 919
 920    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 921        " ": TokenType.SPACE,
 922        "\t": TokenType.SPACE,
 923        "\n": TokenType.BREAK,
 924        "\r": TokenType.BREAK,
 925    }
 926
 927    COMMANDS = {
 928        TokenType.COMMAND,
 929        TokenType.EXECUTE,
 930        TokenType.FETCH,
 931        TokenType.SHOW,
 932        TokenType.RENAME,
 933    }
 934
 935    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 936
 937    # Handle numeric literals like in hive (3L = BIGINT)
 938    NUMERIC_LITERALS: t.Dict[str, str] = {}
 939
 940    COMMENTS = ["--", ("/*", "*/")]
 941
 942    __slots__ = (
 943        "sql",
 944        "size",
 945        "tokens",
 946        "dialect",
 947        "_start",
 948        "_current",
 949        "_line",
 950        "_col",
 951        "_comments",
 952        "_char",
 953        "_end",
 954        "_peek",
 955        "_prev_token_line",
 956        "_rs_dialect_settings",
 957    )
 958
 959    def __init__(self, dialect: DialectType = None) -> None:
 960        from sqlglot.dialects import Dialect
 961
 962        self.dialect = Dialect.get_or_raise(dialect)
 963
 964        if USE_RS_TOKENIZER:
 965            self._rs_dialect_settings = RsTokenizerDialectSettings(
 966                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 967                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 968            )
 969
 970        self.reset()
 971
 972    def reset(self) -> None:
 973        self.sql = ""
 974        self.size = 0
 975        self.tokens: t.List[Token] = []
 976        self._start = 0
 977        self._current = 0
 978        self._line = 1
 979        self._col = 0
 980        self._comments: t.List[str] = []
 981
 982        self._char = ""
 983        self._end = False
 984        self._peek = ""
 985        self._prev_token_line = -1
 986
 987    def tokenize(self, sql: str) -> t.List[Token]:
 988        """Returns a list of tokens corresponding to the SQL string `sql`."""
 989        if USE_RS_TOKENIZER:
 990            return self.tokenize_rs(sql)
 991
 992        self.reset()
 993        self.sql = sql
 994        self.size = len(sql)
 995
 996        try:
 997            self._scan()
 998        except Exception as e:
 999            start = max(self._current - 50, 0)
1000            end = min(self._current + 50, self.size - 1)
1001            context = self.sql[start:end]
1002            raise TokenError(f"Error tokenizing '{context}'") from e
1003
1004        return self.tokens
1005
1006    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
1007        while self.size and not self._end:
1008            current = self._current
1009
1010            # Skip spaces here rather than iteratively calling advance() for performance reasons
1011            while current < self.size:
1012                char = self.sql[current]
1013
1014                if char.isspace() and (char == " " or char == "\t"):
1015                    current += 1
1016                else:
1017                    break
1018
1019            offset = current - self._current if current > self._current else 1
1020
1021            self._start = current
1022            self._advance(offset)
1023
1024            if not self._char.isspace():
1025                if self._char.isdigit():
1026                    self._scan_number()
1027                elif self._char in self._IDENTIFIERS:
1028                    self._scan_identifier(self._IDENTIFIERS[self._char])
1029                else:
1030                    self._scan_keywords()
1031
1032            if until and until():
1033                break
1034
1035        if self.tokens and self._comments:
1036            self.tokens[-1].comments.extend(self._comments)
1037
1038    def _chars(self, size: int) -> str:
1039        if size == 1:
1040            return self._char
1041
1042        start = self._current - 1
1043        end = start + size
1044
1045        return self.sql[start:end] if end <= self.size else ""
1046
1047    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1048        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1049            # Ensures we don't count an extra line if we get a \r\n line break sequence
1050            if not (self._char == "\r" and self._peek == "\n"):
1051                self._col = 1
1052                self._line += 1
1053        else:
1054            self._col += i
1055
1056        self._current += i
1057        self._end = self._current >= self.size
1058        self._char = self.sql[self._current - 1]
1059        self._peek = "" if self._end else self.sql[self._current]
1060
1061        if alnum and self._char.isalnum():
1062            # Here we use local variables instead of attributes for better performance
1063            _col = self._col
1064            _current = self._current
1065            _end = self._end
1066            _peek = self._peek
1067
1068            while _peek.isalnum():
1069                _col += 1
1070                _current += 1
1071                _end = _current >= self.size
1072                _peek = "" if _end else self.sql[_current]
1073
1074            self._col = _col
1075            self._current = _current
1076            self._end = _end
1077            self._peek = _peek
1078            self._char = self.sql[_current - 1]
1079
1080    @property
1081    def _text(self) -> str:
1082        return self.sql[self._start : self._current]
1083
1084    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1085        self._prev_token_line = self._line
1086
1087        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1088            self.tokens[-1].comments.extend(self._comments)
1089            self._comments = []
1090
1091        self.tokens.append(
1092            Token(
1093                token_type,
1094                text=self._text if text is None else text,
1095                line=self._line,
1096                col=self._col,
1097                start=self._start,
1098                end=self._current - 1,
1099                comments=self._comments,
1100            )
1101        )
1102        self._comments = []
1103
1104        # If we have either a semicolon or a begin token before the command's token, we'll parse
1105        # whatever follows the command's token as a string
1106        if (
1107            token_type in self.COMMANDS
1108            and self._peek != ";"
1109            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1110        ):
1111            start = self._current
1112            tokens = len(self.tokens)
1113            self._scan(lambda: self._peek == ";")
1114            self.tokens = self.tokens[:tokens]
1115            text = self.sql[start : self._current].strip()
1116            if text:
1117                self._add(TokenType.STRING, text)
1118
1119    def _scan_keywords(self) -> None:
1120        size = 0
1121        word = None
1122        chars = self._text
1123        char = chars
1124        prev_space = False
1125        skip = False
1126        trie = self._KEYWORD_TRIE
1127        single_token = char in self.SINGLE_TOKENS
1128
1129        while chars:
1130            if skip:
1131                result = TrieResult.PREFIX
1132            else:
1133                result, trie = in_trie(trie, char.upper())
1134
1135            if result == TrieResult.FAILED:
1136                break
1137            if result == TrieResult.EXISTS:
1138                word = chars
1139
1140            end = self._current + size
1141            size += 1
1142
1143            if end < self.size:
1144                char = self.sql[end]
1145                single_token = single_token or char in self.SINGLE_TOKENS
1146                is_space = char.isspace()
1147
1148                if not is_space or not prev_space:
1149                    if is_space:
1150                        char = " "
1151                    chars += char
1152                    prev_space = is_space
1153                    skip = False
1154                else:
1155                    skip = True
1156            else:
1157                char = ""
1158                break
1159
1160        if word:
1161            if self._scan_string(word):
1162                return
1163            if self._scan_comment(word):
1164                return
1165            if prev_space or single_token or not char:
1166                self._advance(size - 1)
1167                word = word.upper()
1168                self._add(self.KEYWORDS[word], text=word)
1169                return
1170
1171        if self._char in self.SINGLE_TOKENS:
1172            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1173            return
1174
1175        self._scan_var()
1176
1177    def _scan_comment(self, comment_start: str) -> bool:
1178        if comment_start not in self._COMMENTS:
1179            return False
1180
1181        comment_start_line = self._line
1182        comment_start_size = len(comment_start)
1183        comment_end = self._COMMENTS[comment_start]
1184
1185        if comment_end:
1186            # Skip the comment's start delimiter
1187            self._advance(comment_start_size)
1188
1189            comment_count = 1
1190            comment_end_size = len(comment_end)
1191
1192            while not self._end:
1193                if self._chars(comment_end_size) == comment_end:
1194                    comment_count -= 1
1195                    if not comment_count:
1196                        break
1197
1198                self._advance(alnum=True)
1199
1200                # Nested comments are allowed by some dialects, e.g. databricks, duckdb, postgres
1201                if (
1202                    self.NESTED_COMMENTS
1203                    and not self._end
1204                    and self._chars(comment_end_size) == comment_start
1205                ):
1206                    self._advance(comment_start_size)
1207                    comment_count += 1
1208
1209            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1210            self._advance(comment_end_size - 1)
1211        else:
1212            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1213                self._advance(alnum=True)
1214            self._comments.append(self._text[comment_start_size:])
1215
1216        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1217        # Multiple consecutive comments are preserved by appending them to the current comments list.
1218        if comment_start_line == self._prev_token_line:
1219            self.tokens[-1].comments.extend(self._comments)
1220            self._comments = []
1221            self._prev_token_line = self._line
1222
1223        return True
1224
1225    def _scan_number(self) -> None:
1226        if self._char == "0":
1227            peek = self._peek.upper()
1228            if peek == "B":
1229                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1230            elif peek == "X":
1231                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1232
1233        decimal = False
1234        scientific = 0
1235
1236        while True:
1237            if self._peek.isdigit():
1238                self._advance()
1239            elif self._peek == "." and not decimal:
1240                if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER:
1241                    return self._add(TokenType.NUMBER)
1242                decimal = True
1243                self._advance()
1244            elif self._peek in ("-", "+") and scientific == 1:
1245                scientific += 1
1246                self._advance()
1247            elif self._peek.upper() == "E" and not scientific:
1248                scientific += 1
1249                self._advance()
1250            elif self._peek.isidentifier():
1251                number_text = self._text
1252                literal = ""
1253
1254                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1255                    literal += self._peek
1256                    self._advance()
1257
1258                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1259
1260                if token_type:
1261                    self._add(TokenType.NUMBER, number_text)
1262                    self._add(TokenType.DCOLON, "::")
1263                    return self._add(token_type, literal)
1264                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1265                    return self._add(TokenType.VAR)
1266
1267                self._advance(-len(literal))
1268                return self._add(TokenType.NUMBER, number_text)
1269            else:
1270                return self._add(TokenType.NUMBER)
1271
1272    def _scan_bits(self) -> None:
1273        self._advance()
1274        value = self._extract_value()
1275        try:
1276            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1277            int(value, 2)
1278            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1279        except ValueError:
1280            self._add(TokenType.IDENTIFIER)
1281
1282    def _scan_hex(self) -> None:
1283        self._advance()
1284        value = self._extract_value()
1285        try:
1286            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1287            int(value, 16)
1288            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1289        except ValueError:
1290            self._add(TokenType.IDENTIFIER)
1291
1292    def _extract_value(self) -> str:
1293        while True:
1294            char = self._peek.strip()
1295            if char and char not in self.SINGLE_TOKENS:
1296                self._advance(alnum=True)
1297            else:
1298                break
1299
1300        return self._text
1301
1302    def _scan_string(self, start: str) -> bool:
1303        base = None
1304        token_type = TokenType.STRING
1305
1306        if start in self._QUOTES:
1307            end = self._QUOTES[start]
1308        elif start in self._FORMAT_STRINGS:
1309            end, token_type = self._FORMAT_STRINGS[start]
1310
1311            if token_type == TokenType.HEX_STRING:
1312                base = 16
1313            elif token_type == TokenType.BIT_STRING:
1314                base = 2
1315            elif token_type == TokenType.HEREDOC_STRING:
1316                self._advance()
1317
1318                if self._char == end:
1319                    tag = ""
1320                else:
1321                    tag = self._extract_string(
1322                        end,
1323                        raw_string=True,
1324                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1325                    )
1326
1327                if tag and self.HEREDOC_TAG_IS_IDENTIFIER and (self._end or not tag.isidentifier()):
1328                    if not self._end:
1329                        self._advance(-1)
1330
1331                    self._advance(-len(tag))
1332                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1333                    return True
1334
1335                end = f"{start}{tag}{end}"
1336        else:
1337            return False
1338
1339        self._advance(len(start))
1340        text = self._extract_string(end, raw_string=token_type == TokenType.RAW_STRING)
1341
1342        if base:
1343            try:
1344                int(text, base)
1345            except Exception:
1346                raise TokenError(
1347                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1348                )
1349
1350        self._add(token_type, text)
1351        return True
1352
1353    def _scan_identifier(self, identifier_end: str) -> None:
1354        self._advance()
1355        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1356        self._add(TokenType.IDENTIFIER, text)
1357
1358    def _scan_var(self) -> None:
1359        while True:
1360            char = self._peek.strip()
1361            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1362                self._advance(alnum=True)
1363            else:
1364                break
1365
1366        self._add(
1367            TokenType.VAR
1368            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1369            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1370        )
1371
1372    def _extract_string(
1373        self,
1374        delimiter: str,
1375        escapes: t.Optional[t.Set[str]] = None,
1376        raw_string: bool = False,
1377        raise_unmatched: bool = True,
1378    ) -> str:
1379        text = ""
1380        delim_size = len(delimiter)
1381        escapes = self._STRING_ESCAPES if escapes is None else escapes
1382
1383        while True:
1384            if (
1385                not raw_string
1386                and self.dialect.UNESCAPED_SEQUENCES
1387                and self._peek
1388                and self._char in self.STRING_ESCAPES
1389            ):
1390                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1391                if unescaped_sequence:
1392                    self._advance(2)
1393                    text += unescaped_sequence
1394                    continue
1395            if (
1396                (self.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS or not raw_string)
1397                and self._char in escapes
1398                and (self._peek == delimiter or self._peek in escapes)
1399                and (self._char not in self._QUOTES or self._char == self._peek)
1400            ):
1401                if self._peek == delimiter:
1402                    text += self._peek
1403                else:
1404                    text += self._char + self._peek
1405
1406                if self._current + 1 < self.size:
1407                    self._advance(2)
1408                else:
1409                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1410            else:
1411                if self._chars(delim_size) == delimiter:
1412                    if delim_size > 1:
1413                        self._advance(delim_size - 1)
1414                    break
1415
1416                if self._end:
1417                    if not raise_unmatched:
1418                        return text + self._char
1419
1420                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1421
1422                current = self._current - 1
1423                self._advance(alnum=True)
1424                text += self.sql[current : self._current - 1]
1425
1426        return text
1427
1428    def tokenize_rs(self, sql: str) -> t.List[Token]:
1429        if not self._RS_TOKENIZER:
1430            raise SqlglotError("Rust tokenizer is not available")
1431
1432        try:
1433            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1434            for token in tokens:
1435                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1436            return tokens
1437        except Exception as e:
1438            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    WAREHOUSE = auto()
 97    STREAMLIT = auto()
 98    VAR = auto()
 99    BIT_STRING = auto()
100    HEX_STRING = auto()
101    BYTE_STRING = auto()
102    NATIONAL_STRING = auto()
103    RAW_STRING = auto()
104    HEREDOC_STRING = auto()
105    UNICODE_STRING = auto()
106
107    # types
108    BIT = auto()
109    BOOLEAN = auto()
110    TINYINT = auto()
111    UTINYINT = auto()
112    SMALLINT = auto()
113    USMALLINT = auto()
114    MEDIUMINT = auto()
115    UMEDIUMINT = auto()
116    INT = auto()
117    UINT = auto()
118    BIGINT = auto()
119    UBIGINT = auto()
120    INT128 = auto()
121    UINT128 = auto()
122    INT256 = auto()
123    UINT256 = auto()
124    FLOAT = auto()
125    DOUBLE = auto()
126    DECIMAL = auto()
127    DECIMAL32 = auto()
128    DECIMAL64 = auto()
129    DECIMAL128 = auto()
130    UDECIMAL = auto()
131    BIGDECIMAL = auto()
132    CHAR = auto()
133    NCHAR = auto()
134    VARCHAR = auto()
135    NVARCHAR = auto()
136    BPCHAR = auto()
137    TEXT = auto()
138    MEDIUMTEXT = auto()
139    LONGTEXT = auto()
140    MEDIUMBLOB = auto()
141    LONGBLOB = auto()
142    TINYBLOB = auto()
143    TINYTEXT = auto()
144    NAME = auto()
145    BINARY = auto()
146    VARBINARY = auto()
147    JSON = auto()
148    JSONB = auto()
149    TIME = auto()
150    TIMETZ = auto()
151    TIMESTAMP = auto()
152    TIMESTAMPTZ = auto()
153    TIMESTAMPLTZ = auto()
154    TIMESTAMPNTZ = auto()
155    TIMESTAMP_S = auto()
156    TIMESTAMP_MS = auto()
157    TIMESTAMP_NS = auto()
158    DATETIME = auto()
159    DATETIME64 = auto()
160    DATE = auto()
161    DATE32 = auto()
162    INT4RANGE = auto()
163    INT4MULTIRANGE = auto()
164    INT8RANGE = auto()
165    INT8MULTIRANGE = auto()
166    NUMRANGE = auto()
167    NUMMULTIRANGE = auto()
168    TSRANGE = auto()
169    TSMULTIRANGE = auto()
170    TSTZRANGE = auto()
171    TSTZMULTIRANGE = auto()
172    DATERANGE = auto()
173    DATEMULTIRANGE = auto()
174    UUID = auto()
175    GEOGRAPHY = auto()
176    NULLABLE = auto()
177    GEOMETRY = auto()
178    HLLSKETCH = auto()
179    HSTORE = auto()
180    SUPER = auto()
181    SERIAL = auto()
182    SMALLSERIAL = auto()
183    BIGSERIAL = auto()
184    XML = auto()
185    YEAR = auto()
186    UNIQUEIDENTIFIER = auto()
187    USERDEFINED = auto()
188    MONEY = auto()
189    SMALLMONEY = auto()
190    ROWVERSION = auto()
191    IMAGE = auto()
192    VARIANT = auto()
193    OBJECT = auto()
194    INET = auto()
195    IPADDRESS = auto()
196    IPPREFIX = auto()
197    IPV4 = auto()
198    IPV6 = auto()
199    ENUM = auto()
200    ENUM8 = auto()
201    ENUM16 = auto()
202    FIXEDSTRING = auto()
203    LOWCARDINALITY = auto()
204    NESTED = auto()
205    AGGREGATEFUNCTION = auto()
206    SIMPLEAGGREGATEFUNCTION = auto()
207    TDIGEST = auto()
208    UNKNOWN = auto()
209    VECTOR = auto()
210
211    # keywords
212    ALIAS = auto()
213    ALTER = auto()
214    ALWAYS = auto()
215    ALL = auto()
216    ANTI = auto()
217    ANY = auto()
218    APPLY = auto()
219    ARRAY = auto()
220    ASC = auto()
221    ASOF = auto()
222    AUTO_INCREMENT = auto()
223    BEGIN = auto()
224    BETWEEN = auto()
225    BULK_COLLECT_INTO = auto()
226    CACHE = auto()
227    CASE = auto()
228    CHARACTER_SET = auto()
229    CLUSTER_BY = auto()
230    COLLATE = auto()
231    COMMAND = auto()
232    COMMENT = auto()
233    COMMIT = auto()
234    CONNECT_BY = auto()
235    CONSTRAINT = auto()
236    COPY = auto()
237    CREATE = auto()
238    CROSS = auto()
239    CUBE = auto()
240    CURRENT_DATE = auto()
241    CURRENT_DATETIME = auto()
242    CURRENT_TIME = auto()
243    CURRENT_TIMESTAMP = auto()
244    CURRENT_USER = auto()
245    DECLARE = auto()
246    DEFAULT = auto()
247    DELETE = auto()
248    DESC = auto()
249    DESCRIBE = auto()
250    DICTIONARY = auto()
251    DISTINCT = auto()
252    DISTRIBUTE_BY = auto()
253    DIV = auto()
254    DROP = auto()
255    ELSE = auto()
256    END = auto()
257    ESCAPE = auto()
258    EXCEPT = auto()
259    EXECUTE = auto()
260    EXISTS = auto()
261    FALSE = auto()
262    FETCH = auto()
263    FILTER = auto()
264    FINAL = auto()
265    FIRST = auto()
266    FOR = auto()
267    FORCE = auto()
268    FOREIGN_KEY = auto()
269    FORMAT = auto()
270    FROM = auto()
271    FULL = auto()
272    FUNCTION = auto()
273    GLOB = auto()
274    GLOBAL = auto()
275    GRANT = auto()
276    GROUP_BY = auto()
277    GROUPING_SETS = auto()
278    HAVING = auto()
279    HINT = auto()
280    IGNORE = auto()
281    ILIKE = auto()
282    ILIKE_ANY = auto()
283    IN = auto()
284    INDEX = auto()
285    INNER = auto()
286    INSERT = auto()
287    INTERSECT = auto()
288    INTERVAL = auto()
289    INTO = auto()
290    INTRODUCER = auto()
291    IRLIKE = auto()
292    IS = auto()
293    ISNULL = auto()
294    JOIN = auto()
295    JOIN_MARKER = auto()
296    KEEP = auto()
297    KEY = auto()
298    KILL = auto()
299    LANGUAGE = auto()
300    LATERAL = auto()
301    LEFT = auto()
302    LIKE = auto()
303    LIKE_ANY = auto()
304    LIMIT = auto()
305    LIST = auto()
306    LOAD = auto()
307    LOCK = auto()
308    MAP = auto()
309    MATCH_CONDITION = auto()
310    MATCH_RECOGNIZE = auto()
311    MEMBER_OF = auto()
312    MERGE = auto()
313    MOD = auto()
314    MODEL = auto()
315    NATURAL = auto()
316    NEXT = auto()
317    NOTNULL = auto()
318    NULL = auto()
319    OBJECT_IDENTIFIER = auto()
320    OFFSET = auto()
321    ON = auto()
322    ONLY = auto()
323    OPERATOR = auto()
324    ORDER_BY = auto()
325    ORDER_SIBLINGS_BY = auto()
326    ORDERED = auto()
327    ORDINALITY = auto()
328    OUTER = auto()
329    OVER = auto()
330    OVERLAPS = auto()
331    OVERWRITE = auto()
332    PARTITION = auto()
333    PARTITION_BY = auto()
334    PERCENT = auto()
335    PIVOT = auto()
336    PLACEHOLDER = auto()
337    POSITIONAL = auto()
338    PRAGMA = auto()
339    PREWHERE = auto()
340    PRIMARY_KEY = auto()
341    PROCEDURE = auto()
342    PROPERTIES = auto()
343    PSEUDO_TYPE = auto()
344    QUALIFY = auto()
345    QUOTE = auto()
346    RANGE = auto()
347    RECURSIVE = auto()
348    REFRESH = auto()
349    RENAME = auto()
350    REPLACE = auto()
351    RETURNING = auto()
352    REFERENCES = auto()
353    RIGHT = auto()
354    RLIKE = auto()
355    ROLLBACK = auto()
356    ROLLUP = auto()
357    ROW = auto()
358    ROWS = auto()
359    SELECT = auto()
360    SEMI = auto()
361    SEPARATOR = auto()
362    SEQUENCE = auto()
363    SERDE_PROPERTIES = auto()
364    SET = auto()
365    SETTINGS = auto()
366    SHOW = auto()
367    SIMILAR_TO = auto()
368    SOME = auto()
369    SORT_BY = auto()
370    START_WITH = auto()
371    STORAGE_INTEGRATION = auto()
372    STRAIGHT_JOIN = auto()
373    STRUCT = auto()
374    SUMMARIZE = auto()
375    TABLE_SAMPLE = auto()
376    TAG = auto()
377    TEMPORARY = auto()
378    TOP = auto()
379    THEN = auto()
380    TRUE = auto()
381    TRUNCATE = auto()
382    UNCACHE = auto()
383    UNION = auto()
384    UNNEST = auto()
385    UNPIVOT = auto()
386    UPDATE = auto()
387    USE = auto()
388    USING = auto()
389    VALUES = auto()
390    VIEW = auto()
391    VOLATILE = auto()
392    WHEN = auto()
393    WHERE = auto()
394    WINDOW = auto()
395    WITH = auto()
396    UNIQUE = auto()
397    VERSION_SNAPSHOT = auto()
398    TIMESTAMP_SNAPSHOT = auto()
399    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'>
WAREHOUSE = <TokenType.WAREHOUSE: 'WAREHOUSE'>
STREAMLIT = <TokenType.STREAMLIT: 'STREAMLIT'>
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'>
DECIMAL32 = <TokenType.DECIMAL32: 'DECIMAL32'>
DECIMAL64 = <TokenType.DECIMAL64: 'DECIMAL64'>
DECIMAL128 = <TokenType.DECIMAL128: 'DECIMAL128'>
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'>
TIMESTAMPNTZ = <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>
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'>
TDIGEST = <TokenType.TDIGEST: 'TDIGEST'>
UNKNOWN = <TokenType.UNKNOWN: 'UNKNOWN'>
VECTOR = <TokenType.VECTOR: 'VECTOR'>
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'>
BULK_COLLECT_INTO = <TokenType.BULK_COLLECT_INTO: 'BULK_COLLECT_INTO'>
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'>
DECLARE = <TokenType.DECLARE: 'DECLARE'>
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'>
GRANT = <TokenType.GRANT: 'GRANT'>
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'>
KEY = <TokenType.KEY: 'KEY'>
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'>
LIST = <TokenType.LIST: 'LIST'>
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'>
RENAME = <TokenType.RENAME: 'RENAME'>
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'>
STRAIGHT_JOIN = <TokenType.STRAIGHT_JOIN: 'STRAIGHT_JOIN'>
STRUCT = <TokenType.STRUCT: 'STRUCT'>
SUMMARIZE = <TokenType.SUMMARIZE: 'SUMMARIZE'>
TABLE_SAMPLE = <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>
TAG = <TokenType.TAG: 'TAG'>
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:
406class Token:
407    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
408
409    @classmethod
410    def number(cls, number: int) -> Token:
411        """Returns a NUMBER token with `number` as its text."""
412        return cls(TokenType.NUMBER, str(number))
413
414    @classmethod
415    def string(cls, string: str) -> Token:
416        """Returns a STRING token with `string` as its text."""
417        return cls(TokenType.STRING, string)
418
419    @classmethod
420    def identifier(cls, identifier: str) -> Token:
421        """Returns an IDENTIFIER token with `identifier` as its text."""
422        return cls(TokenType.IDENTIFIER, identifier)
423
424    @classmethod
425    def var(cls, var: str) -> Token:
426        """Returns an VAR token with `var` as its text."""
427        return cls(TokenType.VAR, var)
428
429    def __init__(
430        self,
431        token_type: TokenType,
432        text: str,
433        line: int = 1,
434        col: int = 1,
435        start: int = 0,
436        end: int = 0,
437        comments: t.Optional[t.List[str]] = None,
438    ) -> None:
439        """Token initializer.
440
441        Args:
442            token_type: The TokenType Enum.
443            text: The text of the token.
444            line: The line that the token ends on.
445            col: The column that the token ends on.
446            start: The start index of the token.
447            end: The ending index of the token.
448            comments: The comments to attach to the token.
449        """
450        self.token_type = token_type
451        self.text = text
452        self.line = line
453        self.col = col
454        self.start = start
455        self.end = end
456        self.comments = [] if comments is None else comments
457
458    def __repr__(self) -> str:
459        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
460        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)
429    def __init__(
430        self,
431        token_type: TokenType,
432        text: str,
433        line: int = 1,
434        col: int = 1,
435        start: int = 0,
436        end: int = 0,
437        comments: t.Optional[t.List[str]] = None,
438    ) -> None:
439        """Token initializer.
440
441        Args:
442            token_type: The TokenType Enum.
443            text: The text of the token.
444            line: The line that the token ends on.
445            col: The column that the token ends on.
446            start: The start index of the token.
447            end: The ending index of the token.
448            comments: The comments to attach to the token.
449        """
450        self.token_type = token_type
451        self.text = text
452        self.line = line
453        self.col = col
454        self.start = start
455        self.end = end
456        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:
409    @classmethod
410    def number(cls, number: int) -> Token:
411        """Returns a NUMBER token with `number` as its text."""
412        return cls(TokenType.NUMBER, str(number))

Returns a NUMBER token with number as its text.

@classmethod
def string(cls, string: str) -> Token:
414    @classmethod
415    def string(cls, string: str) -> Token:
416        """Returns a STRING token with `string` as its text."""
417        return cls(TokenType.STRING, string)

Returns a STRING token with string as its text.

@classmethod
def identifier(cls, identifier: str) -> Token:
419    @classmethod
420    def identifier(cls, identifier: str) -> Token:
421        """Returns an IDENTIFIER token with `identifier` as its text."""
422        return cls(TokenType.IDENTIFIER, identifier)

Returns an IDENTIFIER token with identifier as its text.

@classmethod
def var(cls, var: str) -> Token:
424    @classmethod
425    def var(cls, var: str) -> Token:
426        """Returns an VAR token with `var` as its text."""
427        return cls(TokenType.VAR, var)

Returns an VAR token with var as its text.

token_type
text
line
col
start
end
comments
class Tokenizer:
 563class Tokenizer(metaclass=_Tokenizer):
 564    SINGLE_TOKENS = {
 565        "(": TokenType.L_PAREN,
 566        ")": TokenType.R_PAREN,
 567        "[": TokenType.L_BRACKET,
 568        "]": TokenType.R_BRACKET,
 569        "{": TokenType.L_BRACE,
 570        "}": TokenType.R_BRACE,
 571        "&": TokenType.AMP,
 572        "^": TokenType.CARET,
 573        ":": TokenType.COLON,
 574        ",": TokenType.COMMA,
 575        ".": TokenType.DOT,
 576        "-": TokenType.DASH,
 577        "=": TokenType.EQ,
 578        ">": TokenType.GT,
 579        "<": TokenType.LT,
 580        "%": TokenType.MOD,
 581        "!": TokenType.NOT,
 582        "|": TokenType.PIPE,
 583        "+": TokenType.PLUS,
 584        ";": TokenType.SEMICOLON,
 585        "/": TokenType.SLASH,
 586        "\\": TokenType.BACKSLASH,
 587        "*": TokenType.STAR,
 588        "~": TokenType.TILDA,
 589        "?": TokenType.PLACEHOLDER,
 590        "@": TokenType.PARAMETER,
 591        "#": TokenType.HASH,
 592        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 593        "'": TokenType.UNKNOWN,
 594        "`": TokenType.UNKNOWN,
 595        '"': TokenType.UNKNOWN,
 596    }
 597
 598    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 599    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 600    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 601    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 602    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 603    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 604    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 605    IDENTIFIER_ESCAPES = ['"']
 606    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 607    STRING_ESCAPES = ["'"]
 608    VAR_SINGLE_TOKENS: t.Set[str] = set()
 609
 610    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 611    HEREDOC_TAG_IS_IDENTIFIER = False
 612
 613    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 614    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 615
 616    # Whether string escape characters function as such when placed within raw strings
 617    STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS = True
 618
 619    NESTED_COMMENTS = True
 620
 621    # Autofilled
 622    _COMMENTS: t.Dict[str, str] = {}
 623    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 624    _IDENTIFIERS: t.Dict[str, str] = {}
 625    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 626    _QUOTES: t.Dict[str, str] = {}
 627    _STRING_ESCAPES: t.Set[str] = set()
 628    _KEYWORD_TRIE: t.Dict = {}
 629    _RS_TOKENIZER: t.Optional[t.Any] = None
 630
 631    KEYWORDS: t.Dict[str, TokenType] = {
 632        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 633        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 634        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 635        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 636        "/*+": TokenType.HINT,
 637        "==": TokenType.EQ,
 638        "::": TokenType.DCOLON,
 639        "||": TokenType.DPIPE,
 640        ">=": TokenType.GTE,
 641        "<=": TokenType.LTE,
 642        "<>": TokenType.NEQ,
 643        "!=": TokenType.NEQ,
 644        ":=": TokenType.COLON_EQ,
 645        "<=>": TokenType.NULLSAFE_EQ,
 646        "->": TokenType.ARROW,
 647        "->>": TokenType.DARROW,
 648        "=>": TokenType.FARROW,
 649        "#>": TokenType.HASH_ARROW,
 650        "#>>": TokenType.DHASH_ARROW,
 651        "<->": TokenType.LR_ARROW,
 652        "&&": TokenType.DAMP,
 653        "??": TokenType.DQMARK,
 654        "~~~": TokenType.GLOB,
 655        "~~": TokenType.LIKE,
 656        "ALL": TokenType.ALL,
 657        "ALWAYS": TokenType.ALWAYS,
 658        "AND": TokenType.AND,
 659        "ANTI": TokenType.ANTI,
 660        "ANY": TokenType.ANY,
 661        "ASC": TokenType.ASC,
 662        "AS": TokenType.ALIAS,
 663        "ASOF": TokenType.ASOF,
 664        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 665        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 666        "BEGIN": TokenType.BEGIN,
 667        "BETWEEN": TokenType.BETWEEN,
 668        "CACHE": TokenType.CACHE,
 669        "UNCACHE": TokenType.UNCACHE,
 670        "CASE": TokenType.CASE,
 671        "CHARACTER SET": TokenType.CHARACTER_SET,
 672        "CLUSTER BY": TokenType.CLUSTER_BY,
 673        "COLLATE": TokenType.COLLATE,
 674        "COLUMN": TokenType.COLUMN,
 675        "COMMIT": TokenType.COMMIT,
 676        "CONNECT BY": TokenType.CONNECT_BY,
 677        "CONSTRAINT": TokenType.CONSTRAINT,
 678        "COPY": TokenType.COPY,
 679        "CREATE": TokenType.CREATE,
 680        "CROSS": TokenType.CROSS,
 681        "CUBE": TokenType.CUBE,
 682        "CURRENT_DATE": TokenType.CURRENT_DATE,
 683        "CURRENT_TIME": TokenType.CURRENT_TIME,
 684        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 685        "CURRENT_USER": TokenType.CURRENT_USER,
 686        "DATABASE": TokenType.DATABASE,
 687        "DEFAULT": TokenType.DEFAULT,
 688        "DELETE": TokenType.DELETE,
 689        "DESC": TokenType.DESC,
 690        "DESCRIBE": TokenType.DESCRIBE,
 691        "DISTINCT": TokenType.DISTINCT,
 692        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 693        "DIV": TokenType.DIV,
 694        "DROP": TokenType.DROP,
 695        "ELSE": TokenType.ELSE,
 696        "END": TokenType.END,
 697        "ENUM": TokenType.ENUM,
 698        "ESCAPE": TokenType.ESCAPE,
 699        "EXCEPT": TokenType.EXCEPT,
 700        "EXECUTE": TokenType.EXECUTE,
 701        "EXISTS": TokenType.EXISTS,
 702        "FALSE": TokenType.FALSE,
 703        "FETCH": TokenType.FETCH,
 704        "FILTER": TokenType.FILTER,
 705        "FIRST": TokenType.FIRST,
 706        "FULL": TokenType.FULL,
 707        "FUNCTION": TokenType.FUNCTION,
 708        "FOR": TokenType.FOR,
 709        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 710        "FORMAT": TokenType.FORMAT,
 711        "FROM": TokenType.FROM,
 712        "GEOGRAPHY": TokenType.GEOGRAPHY,
 713        "GEOMETRY": TokenType.GEOMETRY,
 714        "GLOB": TokenType.GLOB,
 715        "GROUP BY": TokenType.GROUP_BY,
 716        "GROUPING SETS": TokenType.GROUPING_SETS,
 717        "HAVING": TokenType.HAVING,
 718        "ILIKE": TokenType.ILIKE,
 719        "IN": TokenType.IN,
 720        "INDEX": TokenType.INDEX,
 721        "INET": TokenType.INET,
 722        "INNER": TokenType.INNER,
 723        "INSERT": TokenType.INSERT,
 724        "INTERVAL": TokenType.INTERVAL,
 725        "INTERSECT": TokenType.INTERSECT,
 726        "INTO": TokenType.INTO,
 727        "IS": TokenType.IS,
 728        "ISNULL": TokenType.ISNULL,
 729        "JOIN": TokenType.JOIN,
 730        "KEEP": TokenType.KEEP,
 731        "KILL": TokenType.KILL,
 732        "LATERAL": TokenType.LATERAL,
 733        "LEFT": TokenType.LEFT,
 734        "LIKE": TokenType.LIKE,
 735        "LIMIT": TokenType.LIMIT,
 736        "LOAD": TokenType.LOAD,
 737        "LOCK": TokenType.LOCK,
 738        "MERGE": TokenType.MERGE,
 739        "NATURAL": TokenType.NATURAL,
 740        "NEXT": TokenType.NEXT,
 741        "NOT": TokenType.NOT,
 742        "NOTNULL": TokenType.NOTNULL,
 743        "NULL": TokenType.NULL,
 744        "OBJECT": TokenType.OBJECT,
 745        "OFFSET": TokenType.OFFSET,
 746        "ON": TokenType.ON,
 747        "OR": TokenType.OR,
 748        "XOR": TokenType.XOR,
 749        "ORDER BY": TokenType.ORDER_BY,
 750        "ORDINALITY": TokenType.ORDINALITY,
 751        "OUTER": TokenType.OUTER,
 752        "OVER": TokenType.OVER,
 753        "OVERLAPS": TokenType.OVERLAPS,
 754        "OVERWRITE": TokenType.OVERWRITE,
 755        "PARTITION": TokenType.PARTITION,
 756        "PARTITION BY": TokenType.PARTITION_BY,
 757        "PARTITIONED BY": TokenType.PARTITION_BY,
 758        "PARTITIONED_BY": TokenType.PARTITION_BY,
 759        "PERCENT": TokenType.PERCENT,
 760        "PIVOT": TokenType.PIVOT,
 761        "PRAGMA": TokenType.PRAGMA,
 762        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 763        "PROCEDURE": TokenType.PROCEDURE,
 764        "QUALIFY": TokenType.QUALIFY,
 765        "RANGE": TokenType.RANGE,
 766        "RECURSIVE": TokenType.RECURSIVE,
 767        "REGEXP": TokenType.RLIKE,
 768        "RENAME": TokenType.RENAME,
 769        "REPLACE": TokenType.REPLACE,
 770        "RETURNING": TokenType.RETURNING,
 771        "REFERENCES": TokenType.REFERENCES,
 772        "RIGHT": TokenType.RIGHT,
 773        "RLIKE": TokenType.RLIKE,
 774        "ROLLBACK": TokenType.ROLLBACK,
 775        "ROLLUP": TokenType.ROLLUP,
 776        "ROW": TokenType.ROW,
 777        "ROWS": TokenType.ROWS,
 778        "SCHEMA": TokenType.SCHEMA,
 779        "SELECT": TokenType.SELECT,
 780        "SEMI": TokenType.SEMI,
 781        "SET": TokenType.SET,
 782        "SETTINGS": TokenType.SETTINGS,
 783        "SHOW": TokenType.SHOW,
 784        "SIMILAR TO": TokenType.SIMILAR_TO,
 785        "SOME": TokenType.SOME,
 786        "SORT BY": TokenType.SORT_BY,
 787        "START WITH": TokenType.START_WITH,
 788        "STRAIGHT_JOIN": TokenType.STRAIGHT_JOIN,
 789        "TABLE": TokenType.TABLE,
 790        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 791        "TEMP": TokenType.TEMPORARY,
 792        "TEMPORARY": TokenType.TEMPORARY,
 793        "THEN": TokenType.THEN,
 794        "TRUE": TokenType.TRUE,
 795        "TRUNCATE": TokenType.TRUNCATE,
 796        "UNION": TokenType.UNION,
 797        "UNKNOWN": TokenType.UNKNOWN,
 798        "UNNEST": TokenType.UNNEST,
 799        "UNPIVOT": TokenType.UNPIVOT,
 800        "UPDATE": TokenType.UPDATE,
 801        "USE": TokenType.USE,
 802        "USING": TokenType.USING,
 803        "UUID": TokenType.UUID,
 804        "VALUES": TokenType.VALUES,
 805        "VIEW": TokenType.VIEW,
 806        "VOLATILE": TokenType.VOLATILE,
 807        "WHEN": TokenType.WHEN,
 808        "WHERE": TokenType.WHERE,
 809        "WINDOW": TokenType.WINDOW,
 810        "WITH": TokenType.WITH,
 811        "APPLY": TokenType.APPLY,
 812        "ARRAY": TokenType.ARRAY,
 813        "BIT": TokenType.BIT,
 814        "BOOL": TokenType.BOOLEAN,
 815        "BOOLEAN": TokenType.BOOLEAN,
 816        "BYTE": TokenType.TINYINT,
 817        "MEDIUMINT": TokenType.MEDIUMINT,
 818        "INT1": TokenType.TINYINT,
 819        "TINYINT": TokenType.TINYINT,
 820        "INT16": TokenType.SMALLINT,
 821        "SHORT": TokenType.SMALLINT,
 822        "SMALLINT": TokenType.SMALLINT,
 823        "INT128": TokenType.INT128,
 824        "HUGEINT": TokenType.INT128,
 825        "UHUGEINT": TokenType.UINT128,
 826        "INT2": TokenType.SMALLINT,
 827        "INTEGER": TokenType.INT,
 828        "INT": TokenType.INT,
 829        "INT4": TokenType.INT,
 830        "INT32": TokenType.INT,
 831        "INT64": TokenType.BIGINT,
 832        "LONG": TokenType.BIGINT,
 833        "BIGINT": TokenType.BIGINT,
 834        "INT8": TokenType.TINYINT,
 835        "UINT": TokenType.UINT,
 836        "DEC": TokenType.DECIMAL,
 837        "DECIMAL": TokenType.DECIMAL,
 838        "DECIMAL32": TokenType.DECIMAL32,
 839        "DECIMAL64": TokenType.DECIMAL64,
 840        "DECIMAL128": TokenType.DECIMAL128,
 841        "BIGDECIMAL": TokenType.BIGDECIMAL,
 842        "BIGNUMERIC": TokenType.BIGDECIMAL,
 843        "LIST": TokenType.LIST,
 844        "MAP": TokenType.MAP,
 845        "NULLABLE": TokenType.NULLABLE,
 846        "NUMBER": TokenType.DECIMAL,
 847        "NUMERIC": TokenType.DECIMAL,
 848        "FIXED": TokenType.DECIMAL,
 849        "REAL": TokenType.FLOAT,
 850        "FLOAT": TokenType.FLOAT,
 851        "FLOAT4": TokenType.FLOAT,
 852        "FLOAT8": TokenType.DOUBLE,
 853        "DOUBLE": TokenType.DOUBLE,
 854        "DOUBLE PRECISION": TokenType.DOUBLE,
 855        "JSON": TokenType.JSON,
 856        "JSONB": TokenType.JSONB,
 857        "CHAR": TokenType.CHAR,
 858        "CHARACTER": TokenType.CHAR,
 859        "NCHAR": TokenType.NCHAR,
 860        "VARCHAR": TokenType.VARCHAR,
 861        "VARCHAR2": TokenType.VARCHAR,
 862        "NVARCHAR": TokenType.NVARCHAR,
 863        "NVARCHAR2": TokenType.NVARCHAR,
 864        "BPCHAR": TokenType.BPCHAR,
 865        "STR": TokenType.TEXT,
 866        "STRING": TokenType.TEXT,
 867        "TEXT": TokenType.TEXT,
 868        "LONGTEXT": TokenType.LONGTEXT,
 869        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 870        "TINYTEXT": TokenType.TINYTEXT,
 871        "CLOB": TokenType.TEXT,
 872        "LONGVARCHAR": TokenType.TEXT,
 873        "BINARY": TokenType.BINARY,
 874        "BLOB": TokenType.VARBINARY,
 875        "LONGBLOB": TokenType.LONGBLOB,
 876        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 877        "TINYBLOB": TokenType.TINYBLOB,
 878        "BYTEA": TokenType.VARBINARY,
 879        "VARBINARY": TokenType.VARBINARY,
 880        "TIME": TokenType.TIME,
 881        "TIMETZ": TokenType.TIMETZ,
 882        "TIMESTAMP": TokenType.TIMESTAMP,
 883        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 884        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 885        "TIMESTAMP_LTZ": TokenType.TIMESTAMPLTZ,
 886        "TIMESTAMPNTZ": TokenType.TIMESTAMPNTZ,
 887        "TIMESTAMP_NTZ": TokenType.TIMESTAMPNTZ,
 888        "DATE": TokenType.DATE,
 889        "DATETIME": TokenType.DATETIME,
 890        "INT4RANGE": TokenType.INT4RANGE,
 891        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 892        "INT8RANGE": TokenType.INT8RANGE,
 893        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 894        "NUMRANGE": TokenType.NUMRANGE,
 895        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 896        "TSRANGE": TokenType.TSRANGE,
 897        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 898        "TSTZRANGE": TokenType.TSTZRANGE,
 899        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 900        "DATERANGE": TokenType.DATERANGE,
 901        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 902        "UNIQUE": TokenType.UNIQUE,
 903        "VECTOR": TokenType.VECTOR,
 904        "STRUCT": TokenType.STRUCT,
 905        "SEQUENCE": TokenType.SEQUENCE,
 906        "VARIANT": TokenType.VARIANT,
 907        "ALTER": TokenType.ALTER,
 908        "ANALYZE": TokenType.COMMAND,
 909        "CALL": TokenType.COMMAND,
 910        "COMMENT": TokenType.COMMENT,
 911        "EXPLAIN": TokenType.COMMAND,
 912        "GRANT": TokenType.GRANT,
 913        "OPTIMIZE": TokenType.COMMAND,
 914        "PREPARE": TokenType.COMMAND,
 915        "VACUUM": TokenType.COMMAND,
 916        "USER-DEFINED": TokenType.USERDEFINED,
 917        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 918        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 919    }
 920
 921    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 922        " ": TokenType.SPACE,
 923        "\t": TokenType.SPACE,
 924        "\n": TokenType.BREAK,
 925        "\r": TokenType.BREAK,
 926    }
 927
 928    COMMANDS = {
 929        TokenType.COMMAND,
 930        TokenType.EXECUTE,
 931        TokenType.FETCH,
 932        TokenType.SHOW,
 933        TokenType.RENAME,
 934    }
 935
 936    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 937
 938    # Handle numeric literals like in hive (3L = BIGINT)
 939    NUMERIC_LITERALS: t.Dict[str, str] = {}
 940
 941    COMMENTS = ["--", ("/*", "*/")]
 942
 943    __slots__ = (
 944        "sql",
 945        "size",
 946        "tokens",
 947        "dialect",
 948        "_start",
 949        "_current",
 950        "_line",
 951        "_col",
 952        "_comments",
 953        "_char",
 954        "_end",
 955        "_peek",
 956        "_prev_token_line",
 957        "_rs_dialect_settings",
 958    )
 959
 960    def __init__(self, dialect: DialectType = None) -> None:
 961        from sqlglot.dialects import Dialect
 962
 963        self.dialect = Dialect.get_or_raise(dialect)
 964
 965        if USE_RS_TOKENIZER:
 966            self._rs_dialect_settings = RsTokenizerDialectSettings(
 967                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 968                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 969            )
 970
 971        self.reset()
 972
 973    def reset(self) -> None:
 974        self.sql = ""
 975        self.size = 0
 976        self.tokens: t.List[Token] = []
 977        self._start = 0
 978        self._current = 0
 979        self._line = 1
 980        self._col = 0
 981        self._comments: t.List[str] = []
 982
 983        self._char = ""
 984        self._end = False
 985        self._peek = ""
 986        self._prev_token_line = -1
 987
 988    def tokenize(self, sql: str) -> t.List[Token]:
 989        """Returns a list of tokens corresponding to the SQL string `sql`."""
 990        if USE_RS_TOKENIZER:
 991            return self.tokenize_rs(sql)
 992
 993        self.reset()
 994        self.sql = sql
 995        self.size = len(sql)
 996
 997        try:
 998            self._scan()
 999        except Exception as e:
1000            start = max(self._current - 50, 0)
1001            end = min(self._current + 50, self.size - 1)
1002            context = self.sql[start:end]
1003            raise TokenError(f"Error tokenizing '{context}'") from e
1004
1005        return self.tokens
1006
1007    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
1008        while self.size and not self._end:
1009            current = self._current
1010
1011            # Skip spaces here rather than iteratively calling advance() for performance reasons
1012            while current < self.size:
1013                char = self.sql[current]
1014
1015                if char.isspace() and (char == " " or char == "\t"):
1016                    current += 1
1017                else:
1018                    break
1019
1020            offset = current - self._current if current > self._current else 1
1021
1022            self._start = current
1023            self._advance(offset)
1024
1025            if not self._char.isspace():
1026                if self._char.isdigit():
1027                    self._scan_number()
1028                elif self._char in self._IDENTIFIERS:
1029                    self._scan_identifier(self._IDENTIFIERS[self._char])
1030                else:
1031                    self._scan_keywords()
1032
1033            if until and until():
1034                break
1035
1036        if self.tokens and self._comments:
1037            self.tokens[-1].comments.extend(self._comments)
1038
1039    def _chars(self, size: int) -> str:
1040        if size == 1:
1041            return self._char
1042
1043        start = self._current - 1
1044        end = start + size
1045
1046        return self.sql[start:end] if end <= self.size else ""
1047
1048    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1049        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1050            # Ensures we don't count an extra line if we get a \r\n line break sequence
1051            if not (self._char == "\r" and self._peek == "\n"):
1052                self._col = 1
1053                self._line += 1
1054        else:
1055            self._col += i
1056
1057        self._current += i
1058        self._end = self._current >= self.size
1059        self._char = self.sql[self._current - 1]
1060        self._peek = "" if self._end else self.sql[self._current]
1061
1062        if alnum and self._char.isalnum():
1063            # Here we use local variables instead of attributes for better performance
1064            _col = self._col
1065            _current = self._current
1066            _end = self._end
1067            _peek = self._peek
1068
1069            while _peek.isalnum():
1070                _col += 1
1071                _current += 1
1072                _end = _current >= self.size
1073                _peek = "" if _end else self.sql[_current]
1074
1075            self._col = _col
1076            self._current = _current
1077            self._end = _end
1078            self._peek = _peek
1079            self._char = self.sql[_current - 1]
1080
1081    @property
1082    def _text(self) -> str:
1083        return self.sql[self._start : self._current]
1084
1085    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1086        self._prev_token_line = self._line
1087
1088        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1089            self.tokens[-1].comments.extend(self._comments)
1090            self._comments = []
1091
1092        self.tokens.append(
1093            Token(
1094                token_type,
1095                text=self._text if text is None else text,
1096                line=self._line,
1097                col=self._col,
1098                start=self._start,
1099                end=self._current - 1,
1100                comments=self._comments,
1101            )
1102        )
1103        self._comments = []
1104
1105        # If we have either a semicolon or a begin token before the command's token, we'll parse
1106        # whatever follows the command's token as a string
1107        if (
1108            token_type in self.COMMANDS
1109            and self._peek != ";"
1110            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1111        ):
1112            start = self._current
1113            tokens = len(self.tokens)
1114            self._scan(lambda: self._peek == ";")
1115            self.tokens = self.tokens[:tokens]
1116            text = self.sql[start : self._current].strip()
1117            if text:
1118                self._add(TokenType.STRING, text)
1119
1120    def _scan_keywords(self) -> None:
1121        size = 0
1122        word = None
1123        chars = self._text
1124        char = chars
1125        prev_space = False
1126        skip = False
1127        trie = self._KEYWORD_TRIE
1128        single_token = char in self.SINGLE_TOKENS
1129
1130        while chars:
1131            if skip:
1132                result = TrieResult.PREFIX
1133            else:
1134                result, trie = in_trie(trie, char.upper())
1135
1136            if result == TrieResult.FAILED:
1137                break
1138            if result == TrieResult.EXISTS:
1139                word = chars
1140
1141            end = self._current + size
1142            size += 1
1143
1144            if end < self.size:
1145                char = self.sql[end]
1146                single_token = single_token or char in self.SINGLE_TOKENS
1147                is_space = char.isspace()
1148
1149                if not is_space or not prev_space:
1150                    if is_space:
1151                        char = " "
1152                    chars += char
1153                    prev_space = is_space
1154                    skip = False
1155                else:
1156                    skip = True
1157            else:
1158                char = ""
1159                break
1160
1161        if word:
1162            if self._scan_string(word):
1163                return
1164            if self._scan_comment(word):
1165                return
1166            if prev_space or single_token or not char:
1167                self._advance(size - 1)
1168                word = word.upper()
1169                self._add(self.KEYWORDS[word], text=word)
1170                return
1171
1172        if self._char in self.SINGLE_TOKENS:
1173            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1174            return
1175
1176        self._scan_var()
1177
1178    def _scan_comment(self, comment_start: str) -> bool:
1179        if comment_start not in self._COMMENTS:
1180            return False
1181
1182        comment_start_line = self._line
1183        comment_start_size = len(comment_start)
1184        comment_end = self._COMMENTS[comment_start]
1185
1186        if comment_end:
1187            # Skip the comment's start delimiter
1188            self._advance(comment_start_size)
1189
1190            comment_count = 1
1191            comment_end_size = len(comment_end)
1192
1193            while not self._end:
1194                if self._chars(comment_end_size) == comment_end:
1195                    comment_count -= 1
1196                    if not comment_count:
1197                        break
1198
1199                self._advance(alnum=True)
1200
1201                # Nested comments are allowed by some dialects, e.g. databricks, duckdb, postgres
1202                if (
1203                    self.NESTED_COMMENTS
1204                    and not self._end
1205                    and self._chars(comment_end_size) == comment_start
1206                ):
1207                    self._advance(comment_start_size)
1208                    comment_count += 1
1209
1210            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1211            self._advance(comment_end_size - 1)
1212        else:
1213            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1214                self._advance(alnum=True)
1215            self._comments.append(self._text[comment_start_size:])
1216
1217        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1218        # Multiple consecutive comments are preserved by appending them to the current comments list.
1219        if comment_start_line == self._prev_token_line:
1220            self.tokens[-1].comments.extend(self._comments)
1221            self._comments = []
1222            self._prev_token_line = self._line
1223
1224        return True
1225
1226    def _scan_number(self) -> None:
1227        if self._char == "0":
1228            peek = self._peek.upper()
1229            if peek == "B":
1230                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1231            elif peek == "X":
1232                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1233
1234        decimal = False
1235        scientific = 0
1236
1237        while True:
1238            if self._peek.isdigit():
1239                self._advance()
1240            elif self._peek == "." and not decimal:
1241                if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER:
1242                    return self._add(TokenType.NUMBER)
1243                decimal = True
1244                self._advance()
1245            elif self._peek in ("-", "+") and scientific == 1:
1246                scientific += 1
1247                self._advance()
1248            elif self._peek.upper() == "E" and not scientific:
1249                scientific += 1
1250                self._advance()
1251            elif self._peek.isidentifier():
1252                number_text = self._text
1253                literal = ""
1254
1255                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1256                    literal += self._peek
1257                    self._advance()
1258
1259                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1260
1261                if token_type:
1262                    self._add(TokenType.NUMBER, number_text)
1263                    self._add(TokenType.DCOLON, "::")
1264                    return self._add(token_type, literal)
1265                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1266                    return self._add(TokenType.VAR)
1267
1268                self._advance(-len(literal))
1269                return self._add(TokenType.NUMBER, number_text)
1270            else:
1271                return self._add(TokenType.NUMBER)
1272
1273    def _scan_bits(self) -> None:
1274        self._advance()
1275        value = self._extract_value()
1276        try:
1277            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1278            int(value, 2)
1279            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1280        except ValueError:
1281            self._add(TokenType.IDENTIFIER)
1282
1283    def _scan_hex(self) -> None:
1284        self._advance()
1285        value = self._extract_value()
1286        try:
1287            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1288            int(value, 16)
1289            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1290        except ValueError:
1291            self._add(TokenType.IDENTIFIER)
1292
1293    def _extract_value(self) -> str:
1294        while True:
1295            char = self._peek.strip()
1296            if char and char not in self.SINGLE_TOKENS:
1297                self._advance(alnum=True)
1298            else:
1299                break
1300
1301        return self._text
1302
1303    def _scan_string(self, start: str) -> bool:
1304        base = None
1305        token_type = TokenType.STRING
1306
1307        if start in self._QUOTES:
1308            end = self._QUOTES[start]
1309        elif start in self._FORMAT_STRINGS:
1310            end, token_type = self._FORMAT_STRINGS[start]
1311
1312            if token_type == TokenType.HEX_STRING:
1313                base = 16
1314            elif token_type == TokenType.BIT_STRING:
1315                base = 2
1316            elif token_type == TokenType.HEREDOC_STRING:
1317                self._advance()
1318
1319                if self._char == end:
1320                    tag = ""
1321                else:
1322                    tag = self._extract_string(
1323                        end,
1324                        raw_string=True,
1325                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1326                    )
1327
1328                if tag and self.HEREDOC_TAG_IS_IDENTIFIER and (self._end or not tag.isidentifier()):
1329                    if not self._end:
1330                        self._advance(-1)
1331
1332                    self._advance(-len(tag))
1333                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1334                    return True
1335
1336                end = f"{start}{tag}{end}"
1337        else:
1338            return False
1339
1340        self._advance(len(start))
1341        text = self._extract_string(end, raw_string=token_type == TokenType.RAW_STRING)
1342
1343        if base:
1344            try:
1345                int(text, base)
1346            except Exception:
1347                raise TokenError(
1348                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1349                )
1350
1351        self._add(token_type, text)
1352        return True
1353
1354    def _scan_identifier(self, identifier_end: str) -> None:
1355        self._advance()
1356        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1357        self._add(TokenType.IDENTIFIER, text)
1358
1359    def _scan_var(self) -> None:
1360        while True:
1361            char = self._peek.strip()
1362            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1363                self._advance(alnum=True)
1364            else:
1365                break
1366
1367        self._add(
1368            TokenType.VAR
1369            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1370            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1371        )
1372
1373    def _extract_string(
1374        self,
1375        delimiter: str,
1376        escapes: t.Optional[t.Set[str]] = None,
1377        raw_string: bool = False,
1378        raise_unmatched: bool = True,
1379    ) -> str:
1380        text = ""
1381        delim_size = len(delimiter)
1382        escapes = self._STRING_ESCAPES if escapes is None else escapes
1383
1384        while True:
1385            if (
1386                not raw_string
1387                and self.dialect.UNESCAPED_SEQUENCES
1388                and self._peek
1389                and self._char in self.STRING_ESCAPES
1390            ):
1391                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1392                if unescaped_sequence:
1393                    self._advance(2)
1394                    text += unescaped_sequence
1395                    continue
1396            if (
1397                (self.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS or not raw_string)
1398                and self._char in escapes
1399                and (self._peek == delimiter or self._peek in escapes)
1400                and (self._char not in self._QUOTES or self._char == self._peek)
1401            ):
1402                if self._peek == delimiter:
1403                    text += self._peek
1404                else:
1405                    text += self._char + self._peek
1406
1407                if self._current + 1 < self.size:
1408                    self._advance(2)
1409                else:
1410                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1411            else:
1412                if self._chars(delim_size) == delimiter:
1413                    if delim_size > 1:
1414                        self._advance(delim_size - 1)
1415                    break
1416
1417                if self._end:
1418                    if not raise_unmatched:
1419                        return text + self._char
1420
1421                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1422
1423                current = self._current - 1
1424                self._advance(alnum=True)
1425                text += self.sql[current : self._current - 1]
1426
1427        return text
1428
1429    def tokenize_rs(self, sql: str) -> t.List[Token]:
1430        if not self._RS_TOKENIZER:
1431            raise SqlglotError("Rust tokenizer is not available")
1432
1433        try:
1434            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1435            for token in tokens:
1436                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1437            return tokens
1438        except Exception as e:
1439            raise TokenError(str(e))
Tokenizer( dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None)
960    def __init__(self, dialect: DialectType = None) -> None:
961        from sqlglot.dialects import Dialect
962
963        self.dialect = Dialect.get_or_raise(dialect)
964
965        if USE_RS_TOKENIZER:
966            self._rs_dialect_settings = RsTokenizerDialectSettings(
967                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
968                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
969            )
970
971        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'>
STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS = True
NESTED_COMMENTS = True
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'>, '~~~': <TokenType.GLOB: 'GLOB'>, '~~': <TokenType.LIKE: 'LIKE'>, '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'>, 'RENAME': <TokenType.RENAME: 'RENAME'>, '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'>, 'STRAIGHT_JOIN': <TokenType.STRAIGHT_JOIN: 'STRAIGHT_JOIN'>, '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'>, 'UHUGEINT': <TokenType.UINT128: 'UINT128'>, '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'>, 'DECIMAL32': <TokenType.DECIMAL32: 'DECIMAL32'>, 'DECIMAL64': <TokenType.DECIMAL64: 'DECIMAL64'>, 'DECIMAL128': <TokenType.DECIMAL128: 'DECIMAL128'>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'LIST': <TokenType.LIST: 'LIST'>, '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'>, 'JSONB': <TokenType.JSONB: 'JSONB'>, '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'>, 'TIMESTAMP_LTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'TIMESTAMPNTZ': <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>, 'TIMESTAMP_NTZ': <TokenType.TIMESTAMPNTZ: 'TIMESTAMPNTZ'>, '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'>, 'VECTOR': <TokenType.VECTOR: 'VECTOR'>, '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.GRANT: 'GRANT'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>}
WHITE_SPACE: Dict[Optional[str], TokenType] = {' ': <TokenType.SPACE: 'SPACE'>, '\t': <TokenType.SPACE: 'SPACE'>, '\n': <TokenType.BREAK: 'BREAK'>, '\r': <TokenType.BREAK: 'BREAK'>}
COMMANDS = {<TokenType.EXECUTE: 'EXECUTE'>, <TokenType.COMMAND: 'COMMAND'>, <TokenType.FETCH: 'FETCH'>, <TokenType.SHOW: 'SHOW'>, <TokenType.RENAME: 'RENAME'>}
COMMAND_PREFIX_TOKENS = {<TokenType.BEGIN: 'BEGIN'>, <TokenType.SEMICOLON: 'SEMICOLON'>}
NUMERIC_LITERALS: Dict[str, str] = {}
COMMENTS = ['--', ('/*', '*/')]
dialect
def reset(self) -> None:
973    def reset(self) -> None:
974        self.sql = ""
975        self.size = 0
976        self.tokens: t.List[Token] = []
977        self._start = 0
978        self._current = 0
979        self._line = 1
980        self._col = 0
981        self._comments: t.List[str] = []
982
983        self._char = ""
984        self._end = False
985        self._peek = ""
986        self._prev_token_line = -1
def tokenize(self, sql: str) -> List[Token]:
 988    def tokenize(self, sql: str) -> t.List[Token]:
 989        """Returns a list of tokens corresponding to the SQL string `sql`."""
 990        if USE_RS_TOKENIZER:
 991            return self.tokenize_rs(sql)
 992
 993        self.reset()
 994        self.sql = sql
 995        self.size = len(sql)
 996
 997        try:
 998            self._scan()
 999        except Exception as e:
1000            start = max(self._current - 50, 0)
1001            end = min(self._current + 50, self.size - 1)
1002            context = self.sql[start:end]
1003            raise TokenError(f"Error tokenizing '{context}'") from e
1004
1005        return self.tokens

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

def tokenize_rs(self, sql: str) -> List[Token]:
1429    def tokenize_rs(self, sql: str) -> t.List[Token]:
1430        if not self._RS_TOKENIZER:
1431            raise SqlglotError("Rust tokenizer is not available")
1432
1433        try:
1434            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1435            for token in tokens:
1436                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1437            return tokens
1438        except Exception as e:
1439            raise TokenError(str(e))
size
sql
tokens