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    CARET_AT = 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    POINT = auto()
 179    RING = auto()
 180    LINESTRING = auto()
 181    MULTILINESTRING = auto()
 182    POLYGON = auto()
 183    MULTIPOLYGON = auto()
 184    HLLSKETCH = auto()
 185    HSTORE = auto()
 186    SUPER = auto()
 187    SERIAL = auto()
 188    SMALLSERIAL = auto()
 189    BIGSERIAL = auto()
 190    XML = auto()
 191    YEAR = auto()
 192    UNIQUEIDENTIFIER = auto()
 193    USERDEFINED = auto()
 194    MONEY = auto()
 195    SMALLMONEY = auto()
 196    ROWVERSION = auto()
 197    IMAGE = auto()
 198    VARIANT = auto()
 199    OBJECT = auto()
 200    INET = auto()
 201    IPADDRESS = auto()
 202    IPPREFIX = auto()
 203    IPV4 = auto()
 204    IPV6 = auto()
 205    ENUM = auto()
 206    ENUM8 = auto()
 207    ENUM16 = auto()
 208    FIXEDSTRING = auto()
 209    LOWCARDINALITY = auto()
 210    NESTED = auto()
 211    AGGREGATEFUNCTION = auto()
 212    SIMPLEAGGREGATEFUNCTION = auto()
 213    TDIGEST = auto()
 214    UNKNOWN = auto()
 215    VECTOR = auto()
 216
 217    # keywords
 218    ALIAS = auto()
 219    ALTER = auto()
 220    ALWAYS = auto()
 221    ALL = auto()
 222    ANTI = auto()
 223    ANY = auto()
 224    APPLY = auto()
 225    ARRAY = auto()
 226    ASC = auto()
 227    ASOF = auto()
 228    AUTO_INCREMENT = auto()
 229    BEGIN = auto()
 230    BETWEEN = auto()
 231    BULK_COLLECT_INTO = auto()
 232    CACHE = auto()
 233    CASE = auto()
 234    CHARACTER_SET = auto()
 235    CLUSTER_BY = auto()
 236    COLLATE = auto()
 237    COMMAND = auto()
 238    COMMENT = auto()
 239    COMMIT = auto()
 240    CONNECT_BY = auto()
 241    CONSTRAINT = auto()
 242    COPY = auto()
 243    CREATE = auto()
 244    CROSS = auto()
 245    CUBE = auto()
 246    CURRENT_DATE = auto()
 247    CURRENT_DATETIME = auto()
 248    CURRENT_TIME = auto()
 249    CURRENT_TIMESTAMP = auto()
 250    CURRENT_USER = auto()
 251    DECLARE = auto()
 252    DEFAULT = auto()
 253    DELETE = auto()
 254    DESC = auto()
 255    DESCRIBE = auto()
 256    DICTIONARY = auto()
 257    DISTINCT = auto()
 258    DISTRIBUTE_BY = auto()
 259    DIV = auto()
 260    DROP = auto()
 261    ELSE = auto()
 262    END = auto()
 263    ESCAPE = auto()
 264    EXCEPT = auto()
 265    EXECUTE = auto()
 266    EXISTS = auto()
 267    FALSE = auto()
 268    FETCH = auto()
 269    FILTER = auto()
 270    FINAL = auto()
 271    FIRST = auto()
 272    FOR = auto()
 273    FORCE = auto()
 274    FOREIGN_KEY = auto()
 275    FORMAT = auto()
 276    FROM = auto()
 277    FULL = auto()
 278    FUNCTION = auto()
 279    GLOB = auto()
 280    GLOBAL = auto()
 281    GRANT = auto()
 282    GROUP_BY = auto()
 283    GROUPING_SETS = auto()
 284    HAVING = auto()
 285    HINT = auto()
 286    IGNORE = auto()
 287    ILIKE = auto()
 288    ILIKE_ANY = auto()
 289    IN = auto()
 290    INDEX = auto()
 291    INNER = auto()
 292    INSERT = auto()
 293    INTERSECT = auto()
 294    INTERVAL = auto()
 295    INTO = auto()
 296    INTRODUCER = auto()
 297    IRLIKE = auto()
 298    IS = auto()
 299    ISNULL = auto()
 300    JOIN = auto()
 301    JOIN_MARKER = auto()
 302    KEEP = auto()
 303    KEY = auto()
 304    KILL = auto()
 305    LANGUAGE = auto()
 306    LATERAL = auto()
 307    LEFT = auto()
 308    LIKE = auto()
 309    LIKE_ANY = auto()
 310    LIMIT = auto()
 311    LIST = auto()
 312    LOAD = auto()
 313    LOCK = auto()
 314    MAP = auto()
 315    MATCH_CONDITION = auto()
 316    MATCH_RECOGNIZE = auto()
 317    MEMBER_OF = auto()
 318    MERGE = auto()
 319    MOD = auto()
 320    MODEL = auto()
 321    NATURAL = auto()
 322    NEXT = auto()
 323    NOTNULL = auto()
 324    NULL = auto()
 325    OBJECT_IDENTIFIER = auto()
 326    OFFSET = auto()
 327    ON = auto()
 328    ONLY = auto()
 329    OPERATOR = auto()
 330    ORDER_BY = auto()
 331    ORDER_SIBLINGS_BY = auto()
 332    ORDERED = auto()
 333    ORDINALITY = auto()
 334    OUTER = auto()
 335    OVER = auto()
 336    OVERLAPS = auto()
 337    OVERWRITE = auto()
 338    PARTITION = auto()
 339    PARTITION_BY = auto()
 340    PERCENT = auto()
 341    PIVOT = auto()
 342    PLACEHOLDER = auto()
 343    POSITIONAL = auto()
 344    PRAGMA = auto()
 345    PREWHERE = auto()
 346    PRIMARY_KEY = auto()
 347    PROCEDURE = auto()
 348    PROPERTIES = auto()
 349    PSEUDO_TYPE = auto()
 350    QUALIFY = auto()
 351    QUOTE = auto()
 352    RANGE = auto()
 353    RECURSIVE = auto()
 354    REFRESH = auto()
 355    RENAME = auto()
 356    REPLACE = auto()
 357    RETURNING = auto()
 358    REFERENCES = auto()
 359    RIGHT = auto()
 360    RLIKE = auto()
 361    ROLLBACK = auto()
 362    ROLLUP = auto()
 363    ROW = auto()
 364    ROWS = auto()
 365    SELECT = auto()
 366    SEMI = auto()
 367    SEPARATOR = auto()
 368    SEQUENCE = auto()
 369    SERDE_PROPERTIES = auto()
 370    SET = auto()
 371    SETTINGS = auto()
 372    SHOW = auto()
 373    SIMILAR_TO = auto()
 374    SOME = auto()
 375    SORT_BY = auto()
 376    START_WITH = auto()
 377    STORAGE_INTEGRATION = auto()
 378    STRAIGHT_JOIN = auto()
 379    STRUCT = auto()
 380    SUMMARIZE = auto()
 381    TABLE_SAMPLE = auto()
 382    TAG = auto()
 383    TEMPORARY = auto()
 384    TOP = auto()
 385    THEN = auto()
 386    TRUE = auto()
 387    TRUNCATE = auto()
 388    UNCACHE = auto()
 389    UNION = auto()
 390    UNNEST = auto()
 391    UNPIVOT = auto()
 392    UPDATE = auto()
 393    USE = auto()
 394    USING = auto()
 395    VALUES = auto()
 396    VIEW = auto()
 397    VOLATILE = auto()
 398    WHEN = auto()
 399    WHERE = auto()
 400    WINDOW = auto()
 401    WITH = auto()
 402    UNIQUE = auto()
 403    VERSION_SNAPSHOT = auto()
 404    TIMESTAMP_SNAPSHOT = auto()
 405    OPTION = auto()
 406
 407
 408_ALL_TOKEN_TYPES = list(TokenType)
 409_TOKEN_TYPE_TO_INDEX = {token_type: i for i, token_type in enumerate(_ALL_TOKEN_TYPES)}
 410
 411
 412class Token:
 413    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
 414
 415    @classmethod
 416    def number(cls, number: int) -> Token:
 417        """Returns a NUMBER token with `number` as its text."""
 418        return cls(TokenType.NUMBER, str(number))
 419
 420    @classmethod
 421    def string(cls, string: str) -> Token:
 422        """Returns a STRING token with `string` as its text."""
 423        return cls(TokenType.STRING, string)
 424
 425    @classmethod
 426    def identifier(cls, identifier: str) -> Token:
 427        """Returns an IDENTIFIER token with `identifier` as its text."""
 428        return cls(TokenType.IDENTIFIER, identifier)
 429
 430    @classmethod
 431    def var(cls, var: str) -> Token:
 432        """Returns an VAR token with `var` as its text."""
 433        return cls(TokenType.VAR, var)
 434
 435    def __init__(
 436        self,
 437        token_type: TokenType,
 438        text: str,
 439        line: int = 1,
 440        col: int = 1,
 441        start: int = 0,
 442        end: int = 0,
 443        comments: t.Optional[t.List[str]] = None,
 444    ) -> None:
 445        """Token initializer.
 446
 447        Args:
 448            token_type: The TokenType Enum.
 449            text: The text of the token.
 450            line: The line that the token ends on.
 451            col: The column that the token ends on.
 452            start: The start index of the token.
 453            end: The ending index of the token.
 454            comments: The comments to attach to the token.
 455        """
 456        self.token_type = token_type
 457        self.text = text
 458        self.line = line
 459        self.col = col
 460        self.start = start
 461        self.end = end
 462        self.comments = [] if comments is None else comments
 463
 464    def __repr__(self) -> str:
 465        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
 466        return f"<Token {attributes}>"
 467
 468
 469class _Tokenizer(type):
 470    def __new__(cls, clsname, bases, attrs):
 471        klass = super().__new__(cls, clsname, bases, attrs)
 472
 473        def _convert_quotes(arr: t.List[str | t.Tuple[str, str]]) -> t.Dict[str, str]:
 474            return dict(
 475                (item, item) if isinstance(item, str) else (item[0], item[1]) for item in arr
 476            )
 477
 478        def _quotes_to_format(
 479            token_type: TokenType, arr: t.List[str | t.Tuple[str, str]]
 480        ) -> t.Dict[str, t.Tuple[str, TokenType]]:
 481            return {k: (v, token_type) for k, v in _convert_quotes(arr).items()}
 482
 483        klass._QUOTES = _convert_quotes(klass.QUOTES)
 484        klass._IDENTIFIERS = _convert_quotes(klass.IDENTIFIERS)
 485
 486        klass._FORMAT_STRINGS = {
 487            **{
 488                p + s: (e, TokenType.NATIONAL_STRING)
 489                for s, e in klass._QUOTES.items()
 490                for p in ("n", "N")
 491            },
 492            **_quotes_to_format(TokenType.BIT_STRING, klass.BIT_STRINGS),
 493            **_quotes_to_format(TokenType.BYTE_STRING, klass.BYTE_STRINGS),
 494            **_quotes_to_format(TokenType.HEX_STRING, klass.HEX_STRINGS),
 495            **_quotes_to_format(TokenType.RAW_STRING, klass.RAW_STRINGS),
 496            **_quotes_to_format(TokenType.HEREDOC_STRING, klass.HEREDOC_STRINGS),
 497            **_quotes_to_format(TokenType.UNICODE_STRING, klass.UNICODE_STRINGS),
 498        }
 499
 500        klass._STRING_ESCAPES = set(klass.STRING_ESCAPES)
 501        klass._IDENTIFIER_ESCAPES = set(klass.IDENTIFIER_ESCAPES)
 502        klass._COMMENTS = {
 503            **dict(
 504                (comment, None) if isinstance(comment, str) else (comment[0], comment[1])
 505                for comment in klass.COMMENTS
 506            ),
 507            "{#": "#}",  # Ensure Jinja comments are tokenized correctly in all dialects
 508        }
 509
 510        klass._KEYWORD_TRIE = new_trie(
 511            key.upper()
 512            for key in (
 513                *klass.KEYWORDS,
 514                *klass._COMMENTS,
 515                *klass._QUOTES,
 516                *klass._FORMAT_STRINGS,
 517            )
 518            if " " in key or any(single in key for single in klass.SINGLE_TOKENS)
 519        )
 520
 521        if USE_RS_TOKENIZER:
 522            settings = RsTokenizerSettings(
 523                white_space={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.WHITE_SPACE.items()},
 524                single_tokens={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.SINGLE_TOKENS.items()},
 525                keywords={k: _TOKEN_TYPE_TO_INDEX[v] for k, v in klass.KEYWORDS.items()},
 526                numeric_literals=klass.NUMERIC_LITERALS,
 527                identifiers=klass._IDENTIFIERS,
 528                identifier_escapes=klass._IDENTIFIER_ESCAPES,
 529                string_escapes=klass._STRING_ESCAPES,
 530                quotes=klass._QUOTES,
 531                format_strings={
 532                    k: (v1, _TOKEN_TYPE_TO_INDEX[v2])
 533                    for k, (v1, v2) in klass._FORMAT_STRINGS.items()
 534                },
 535                has_bit_strings=bool(klass.BIT_STRINGS),
 536                has_hex_strings=bool(klass.HEX_STRINGS),
 537                comments=klass._COMMENTS,
 538                var_single_tokens=klass.VAR_SINGLE_TOKENS,
 539                commands={_TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMANDS},
 540                command_prefix_tokens={
 541                    _TOKEN_TYPE_TO_INDEX[v] for v in klass.COMMAND_PREFIX_TOKENS
 542                },
 543                heredoc_tag_is_identifier=klass.HEREDOC_TAG_IS_IDENTIFIER,
 544                string_escapes_allowed_in_raw_strings=klass.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS,
 545                nested_comments=klass.NESTED_COMMENTS,
 546            )
 547            token_types = RsTokenTypeSettings(
 548                bit_string=_TOKEN_TYPE_TO_INDEX[TokenType.BIT_STRING],
 549                break_=_TOKEN_TYPE_TO_INDEX[TokenType.BREAK],
 550                dcolon=_TOKEN_TYPE_TO_INDEX[TokenType.DCOLON],
 551                heredoc_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEREDOC_STRING],
 552                raw_string=_TOKEN_TYPE_TO_INDEX[TokenType.RAW_STRING],
 553                hex_string=_TOKEN_TYPE_TO_INDEX[TokenType.HEX_STRING],
 554                identifier=_TOKEN_TYPE_TO_INDEX[TokenType.IDENTIFIER],
 555                number=_TOKEN_TYPE_TO_INDEX[TokenType.NUMBER],
 556                parameter=_TOKEN_TYPE_TO_INDEX[TokenType.PARAMETER],
 557                semicolon=_TOKEN_TYPE_TO_INDEX[TokenType.SEMICOLON],
 558                string=_TOKEN_TYPE_TO_INDEX[TokenType.STRING],
 559                var=_TOKEN_TYPE_TO_INDEX[TokenType.VAR],
 560                heredoc_string_alternative=_TOKEN_TYPE_TO_INDEX[klass.HEREDOC_STRING_ALTERNATIVE],
 561            )
 562            klass._RS_TOKENIZER = RsTokenizer(settings, token_types)
 563        else:
 564            klass._RS_TOKENIZER = None
 565
 566        return klass
 567
 568
 569class Tokenizer(metaclass=_Tokenizer):
 570    SINGLE_TOKENS = {
 571        "(": TokenType.L_PAREN,
 572        ")": TokenType.R_PAREN,
 573        "[": TokenType.L_BRACKET,
 574        "]": TokenType.R_BRACKET,
 575        "{": TokenType.L_BRACE,
 576        "}": TokenType.R_BRACE,
 577        "&": TokenType.AMP,
 578        "^": TokenType.CARET,
 579        ":": TokenType.COLON,
 580        ",": TokenType.COMMA,
 581        ".": TokenType.DOT,
 582        "-": TokenType.DASH,
 583        "=": TokenType.EQ,
 584        ">": TokenType.GT,
 585        "<": TokenType.LT,
 586        "%": TokenType.MOD,
 587        "!": TokenType.NOT,
 588        "|": TokenType.PIPE,
 589        "+": TokenType.PLUS,
 590        ";": TokenType.SEMICOLON,
 591        "/": TokenType.SLASH,
 592        "\\": TokenType.BACKSLASH,
 593        "*": TokenType.STAR,
 594        "~": TokenType.TILDA,
 595        "?": TokenType.PLACEHOLDER,
 596        "@": TokenType.PARAMETER,
 597        "#": TokenType.HASH,
 598        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 599        "'": TokenType.UNKNOWN,
 600        "`": TokenType.UNKNOWN,
 601        '"': TokenType.UNKNOWN,
 602    }
 603
 604    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 605    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 606    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 607    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 608    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 609    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 610    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 611    IDENTIFIER_ESCAPES = ['"']
 612    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 613    STRING_ESCAPES = ["'"]
 614    VAR_SINGLE_TOKENS: t.Set[str] = set()
 615
 616    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 617    HEREDOC_TAG_IS_IDENTIFIER = False
 618
 619    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 620    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 621
 622    # Whether string escape characters function as such when placed within raw strings
 623    STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS = True
 624
 625    NESTED_COMMENTS = True
 626
 627    # Autofilled
 628    _COMMENTS: t.Dict[str, str] = {}
 629    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 630    _IDENTIFIERS: t.Dict[str, str] = {}
 631    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 632    _QUOTES: t.Dict[str, str] = {}
 633    _STRING_ESCAPES: t.Set[str] = set()
 634    _KEYWORD_TRIE: t.Dict = {}
 635    _RS_TOKENIZER: t.Optional[t.Any] = None
 636
 637    KEYWORDS: t.Dict[str, TokenType] = {
 638        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 639        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 640        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 641        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 642        "/*+": TokenType.HINT,
 643        "==": TokenType.EQ,
 644        "::": TokenType.DCOLON,
 645        "||": TokenType.DPIPE,
 646        ">=": TokenType.GTE,
 647        "<=": TokenType.LTE,
 648        "<>": TokenType.NEQ,
 649        "!=": TokenType.NEQ,
 650        ":=": TokenType.COLON_EQ,
 651        "<=>": TokenType.NULLSAFE_EQ,
 652        "->": TokenType.ARROW,
 653        "->>": TokenType.DARROW,
 654        "=>": TokenType.FARROW,
 655        "#>": TokenType.HASH_ARROW,
 656        "#>>": TokenType.DHASH_ARROW,
 657        "<->": TokenType.LR_ARROW,
 658        "&&": TokenType.DAMP,
 659        "??": TokenType.DQMARK,
 660        "~~~": TokenType.GLOB,
 661        "~~": TokenType.LIKE,
 662        "~~*": TokenType.ILIKE,
 663        "~*": TokenType.IRLIKE,
 664        "ALL": TokenType.ALL,
 665        "ALWAYS": TokenType.ALWAYS,
 666        "AND": TokenType.AND,
 667        "ANTI": TokenType.ANTI,
 668        "ANY": TokenType.ANY,
 669        "ASC": TokenType.ASC,
 670        "AS": TokenType.ALIAS,
 671        "ASOF": TokenType.ASOF,
 672        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 673        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 674        "BEGIN": TokenType.BEGIN,
 675        "BETWEEN": TokenType.BETWEEN,
 676        "CACHE": TokenType.CACHE,
 677        "UNCACHE": TokenType.UNCACHE,
 678        "CASE": TokenType.CASE,
 679        "CHARACTER SET": TokenType.CHARACTER_SET,
 680        "CLUSTER BY": TokenType.CLUSTER_BY,
 681        "COLLATE": TokenType.COLLATE,
 682        "COLUMN": TokenType.COLUMN,
 683        "COMMIT": TokenType.COMMIT,
 684        "CONNECT BY": TokenType.CONNECT_BY,
 685        "CONSTRAINT": TokenType.CONSTRAINT,
 686        "COPY": TokenType.COPY,
 687        "CREATE": TokenType.CREATE,
 688        "CROSS": TokenType.CROSS,
 689        "CUBE": TokenType.CUBE,
 690        "CURRENT_DATE": TokenType.CURRENT_DATE,
 691        "CURRENT_TIME": TokenType.CURRENT_TIME,
 692        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 693        "CURRENT_USER": TokenType.CURRENT_USER,
 694        "DATABASE": TokenType.DATABASE,
 695        "DEFAULT": TokenType.DEFAULT,
 696        "DELETE": TokenType.DELETE,
 697        "DESC": TokenType.DESC,
 698        "DESCRIBE": TokenType.DESCRIBE,
 699        "DISTINCT": TokenType.DISTINCT,
 700        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 701        "DIV": TokenType.DIV,
 702        "DROP": TokenType.DROP,
 703        "ELSE": TokenType.ELSE,
 704        "END": TokenType.END,
 705        "ENUM": TokenType.ENUM,
 706        "ESCAPE": TokenType.ESCAPE,
 707        "EXCEPT": TokenType.EXCEPT,
 708        "EXECUTE": TokenType.EXECUTE,
 709        "EXISTS": TokenType.EXISTS,
 710        "FALSE": TokenType.FALSE,
 711        "FETCH": TokenType.FETCH,
 712        "FILTER": TokenType.FILTER,
 713        "FIRST": TokenType.FIRST,
 714        "FULL": TokenType.FULL,
 715        "FUNCTION": TokenType.FUNCTION,
 716        "FOR": TokenType.FOR,
 717        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 718        "FORMAT": TokenType.FORMAT,
 719        "FROM": TokenType.FROM,
 720        "GEOGRAPHY": TokenType.GEOGRAPHY,
 721        "GEOMETRY": TokenType.GEOMETRY,
 722        "GLOB": TokenType.GLOB,
 723        "GROUP BY": TokenType.GROUP_BY,
 724        "GROUPING SETS": TokenType.GROUPING_SETS,
 725        "HAVING": TokenType.HAVING,
 726        "ILIKE": TokenType.ILIKE,
 727        "IN": TokenType.IN,
 728        "INDEX": TokenType.INDEX,
 729        "INET": TokenType.INET,
 730        "INNER": TokenType.INNER,
 731        "INSERT": TokenType.INSERT,
 732        "INTERVAL": TokenType.INTERVAL,
 733        "INTERSECT": TokenType.INTERSECT,
 734        "INTO": TokenType.INTO,
 735        "IS": TokenType.IS,
 736        "ISNULL": TokenType.ISNULL,
 737        "JOIN": TokenType.JOIN,
 738        "KEEP": TokenType.KEEP,
 739        "KILL": TokenType.KILL,
 740        "LATERAL": TokenType.LATERAL,
 741        "LEFT": TokenType.LEFT,
 742        "LIKE": TokenType.LIKE,
 743        "LIMIT": TokenType.LIMIT,
 744        "LOAD": TokenType.LOAD,
 745        "LOCK": TokenType.LOCK,
 746        "MERGE": TokenType.MERGE,
 747        "NATURAL": TokenType.NATURAL,
 748        "NEXT": TokenType.NEXT,
 749        "NOT": TokenType.NOT,
 750        "NOTNULL": TokenType.NOTNULL,
 751        "NULL": TokenType.NULL,
 752        "OBJECT": TokenType.OBJECT,
 753        "OFFSET": TokenType.OFFSET,
 754        "ON": TokenType.ON,
 755        "OR": TokenType.OR,
 756        "XOR": TokenType.XOR,
 757        "ORDER BY": TokenType.ORDER_BY,
 758        "ORDINALITY": TokenType.ORDINALITY,
 759        "OUTER": TokenType.OUTER,
 760        "OVER": TokenType.OVER,
 761        "OVERLAPS": TokenType.OVERLAPS,
 762        "OVERWRITE": TokenType.OVERWRITE,
 763        "PARTITION": TokenType.PARTITION,
 764        "PARTITION BY": TokenType.PARTITION_BY,
 765        "PARTITIONED BY": TokenType.PARTITION_BY,
 766        "PARTITIONED_BY": TokenType.PARTITION_BY,
 767        "PERCENT": TokenType.PERCENT,
 768        "PIVOT": TokenType.PIVOT,
 769        "PRAGMA": TokenType.PRAGMA,
 770        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 771        "PROCEDURE": TokenType.PROCEDURE,
 772        "QUALIFY": TokenType.QUALIFY,
 773        "RANGE": TokenType.RANGE,
 774        "RECURSIVE": TokenType.RECURSIVE,
 775        "REGEXP": TokenType.RLIKE,
 776        "RENAME": TokenType.RENAME,
 777        "REPLACE": TokenType.REPLACE,
 778        "RETURNING": TokenType.RETURNING,
 779        "REFERENCES": TokenType.REFERENCES,
 780        "RIGHT": TokenType.RIGHT,
 781        "RLIKE": TokenType.RLIKE,
 782        "ROLLBACK": TokenType.ROLLBACK,
 783        "ROLLUP": TokenType.ROLLUP,
 784        "ROW": TokenType.ROW,
 785        "ROWS": TokenType.ROWS,
 786        "SCHEMA": TokenType.SCHEMA,
 787        "SELECT": TokenType.SELECT,
 788        "SEMI": TokenType.SEMI,
 789        "SET": TokenType.SET,
 790        "SETTINGS": TokenType.SETTINGS,
 791        "SHOW": TokenType.SHOW,
 792        "SIMILAR TO": TokenType.SIMILAR_TO,
 793        "SOME": TokenType.SOME,
 794        "SORT BY": TokenType.SORT_BY,
 795        "START WITH": TokenType.START_WITH,
 796        "STRAIGHT_JOIN": TokenType.STRAIGHT_JOIN,
 797        "TABLE": TokenType.TABLE,
 798        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 799        "TEMP": TokenType.TEMPORARY,
 800        "TEMPORARY": TokenType.TEMPORARY,
 801        "THEN": TokenType.THEN,
 802        "TRUE": TokenType.TRUE,
 803        "TRUNCATE": TokenType.TRUNCATE,
 804        "UNION": TokenType.UNION,
 805        "UNKNOWN": TokenType.UNKNOWN,
 806        "UNNEST": TokenType.UNNEST,
 807        "UNPIVOT": TokenType.UNPIVOT,
 808        "UPDATE": TokenType.UPDATE,
 809        "USE": TokenType.USE,
 810        "USING": TokenType.USING,
 811        "UUID": TokenType.UUID,
 812        "VALUES": TokenType.VALUES,
 813        "VIEW": TokenType.VIEW,
 814        "VOLATILE": TokenType.VOLATILE,
 815        "WHEN": TokenType.WHEN,
 816        "WHERE": TokenType.WHERE,
 817        "WINDOW": TokenType.WINDOW,
 818        "WITH": TokenType.WITH,
 819        "APPLY": TokenType.APPLY,
 820        "ARRAY": TokenType.ARRAY,
 821        "BIT": TokenType.BIT,
 822        "BOOL": TokenType.BOOLEAN,
 823        "BOOLEAN": TokenType.BOOLEAN,
 824        "BYTE": TokenType.TINYINT,
 825        "MEDIUMINT": TokenType.MEDIUMINT,
 826        "INT1": TokenType.TINYINT,
 827        "TINYINT": TokenType.TINYINT,
 828        "INT16": TokenType.SMALLINT,
 829        "SHORT": TokenType.SMALLINT,
 830        "SMALLINT": TokenType.SMALLINT,
 831        "INT128": TokenType.INT128,
 832        "HUGEINT": TokenType.INT128,
 833        "UHUGEINT": TokenType.UINT128,
 834        "INT2": TokenType.SMALLINT,
 835        "INTEGER": TokenType.INT,
 836        "INT": TokenType.INT,
 837        "INT4": TokenType.INT,
 838        "INT32": TokenType.INT,
 839        "INT64": TokenType.BIGINT,
 840        "LONG": TokenType.BIGINT,
 841        "BIGINT": TokenType.BIGINT,
 842        "INT8": TokenType.TINYINT,
 843        "UINT": TokenType.UINT,
 844        "DEC": TokenType.DECIMAL,
 845        "DECIMAL": TokenType.DECIMAL,
 846        "DECIMAL32": TokenType.DECIMAL32,
 847        "DECIMAL64": TokenType.DECIMAL64,
 848        "DECIMAL128": TokenType.DECIMAL128,
 849        "BIGDECIMAL": TokenType.BIGDECIMAL,
 850        "BIGNUMERIC": TokenType.BIGDECIMAL,
 851        "LIST": TokenType.LIST,
 852        "MAP": TokenType.MAP,
 853        "NULLABLE": TokenType.NULLABLE,
 854        "NUMBER": TokenType.DECIMAL,
 855        "NUMERIC": TokenType.DECIMAL,
 856        "FIXED": TokenType.DECIMAL,
 857        "REAL": TokenType.FLOAT,
 858        "FLOAT": TokenType.FLOAT,
 859        "FLOAT4": TokenType.FLOAT,
 860        "FLOAT8": TokenType.DOUBLE,
 861        "DOUBLE": TokenType.DOUBLE,
 862        "DOUBLE PRECISION": TokenType.DOUBLE,
 863        "JSON": TokenType.JSON,
 864        "JSONB": TokenType.JSONB,
 865        "CHAR": TokenType.CHAR,
 866        "CHARACTER": TokenType.CHAR,
 867        "NCHAR": TokenType.NCHAR,
 868        "VARCHAR": TokenType.VARCHAR,
 869        "VARCHAR2": TokenType.VARCHAR,
 870        "NVARCHAR": TokenType.NVARCHAR,
 871        "NVARCHAR2": TokenType.NVARCHAR,
 872        "BPCHAR": TokenType.BPCHAR,
 873        "STR": TokenType.TEXT,
 874        "STRING": TokenType.TEXT,
 875        "TEXT": TokenType.TEXT,
 876        "LONGTEXT": TokenType.LONGTEXT,
 877        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 878        "TINYTEXT": TokenType.TINYTEXT,
 879        "CLOB": TokenType.TEXT,
 880        "LONGVARCHAR": TokenType.TEXT,
 881        "BINARY": TokenType.BINARY,
 882        "BLOB": TokenType.VARBINARY,
 883        "LONGBLOB": TokenType.LONGBLOB,
 884        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 885        "TINYBLOB": TokenType.TINYBLOB,
 886        "BYTEA": TokenType.VARBINARY,
 887        "VARBINARY": TokenType.VARBINARY,
 888        "TIME": TokenType.TIME,
 889        "TIMETZ": TokenType.TIMETZ,
 890        "TIMESTAMP": TokenType.TIMESTAMP,
 891        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 892        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 893        "TIMESTAMP_LTZ": TokenType.TIMESTAMPLTZ,
 894        "TIMESTAMPNTZ": TokenType.TIMESTAMPNTZ,
 895        "TIMESTAMP_NTZ": TokenType.TIMESTAMPNTZ,
 896        "DATE": TokenType.DATE,
 897        "DATETIME": TokenType.DATETIME,
 898        "INT4RANGE": TokenType.INT4RANGE,
 899        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 900        "INT8RANGE": TokenType.INT8RANGE,
 901        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 902        "NUMRANGE": TokenType.NUMRANGE,
 903        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 904        "TSRANGE": TokenType.TSRANGE,
 905        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 906        "TSTZRANGE": TokenType.TSTZRANGE,
 907        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 908        "DATERANGE": TokenType.DATERANGE,
 909        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 910        "UNIQUE": TokenType.UNIQUE,
 911        "VECTOR": TokenType.VECTOR,
 912        "STRUCT": TokenType.STRUCT,
 913        "SEQUENCE": TokenType.SEQUENCE,
 914        "VARIANT": TokenType.VARIANT,
 915        "ALTER": TokenType.ALTER,
 916        "ANALYZE": TokenType.COMMAND,
 917        "CALL": TokenType.COMMAND,
 918        "COMMENT": TokenType.COMMENT,
 919        "EXPLAIN": TokenType.COMMAND,
 920        "GRANT": TokenType.GRANT,
 921        "OPTIMIZE": TokenType.COMMAND,
 922        "PREPARE": TokenType.COMMAND,
 923        "VACUUM": TokenType.COMMAND,
 924        "USER-DEFINED": TokenType.USERDEFINED,
 925        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 926        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 927    }
 928
 929    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 930        " ": TokenType.SPACE,
 931        "\t": TokenType.SPACE,
 932        "\n": TokenType.BREAK,
 933        "\r": TokenType.BREAK,
 934    }
 935
 936    COMMANDS = {
 937        TokenType.COMMAND,
 938        TokenType.EXECUTE,
 939        TokenType.FETCH,
 940        TokenType.SHOW,
 941        TokenType.RENAME,
 942    }
 943
 944    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 945
 946    # Handle numeric literals like in hive (3L = BIGINT)
 947    NUMERIC_LITERALS: t.Dict[str, str] = {}
 948
 949    COMMENTS = ["--", ("/*", "*/")]
 950
 951    __slots__ = (
 952        "sql",
 953        "size",
 954        "tokens",
 955        "dialect",
 956        "_start",
 957        "_current",
 958        "_line",
 959        "_col",
 960        "_comments",
 961        "_char",
 962        "_end",
 963        "_peek",
 964        "_prev_token_line",
 965        "_rs_dialect_settings",
 966    )
 967
 968    def __init__(self, dialect: DialectType = None) -> None:
 969        from sqlglot.dialects import Dialect
 970
 971        self.dialect = Dialect.get_or_raise(dialect)
 972
 973        if USE_RS_TOKENIZER:
 974            self._rs_dialect_settings = RsTokenizerDialectSettings(
 975                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 976                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 977            )
 978
 979        self.reset()
 980
 981    def reset(self) -> None:
 982        self.sql = ""
 983        self.size = 0
 984        self.tokens: t.List[Token] = []
 985        self._start = 0
 986        self._current = 0
 987        self._line = 1
 988        self._col = 0
 989        self._comments: t.List[str] = []
 990
 991        self._char = ""
 992        self._end = False
 993        self._peek = ""
 994        self._prev_token_line = -1
 995
 996    def tokenize(self, sql: str) -> t.List[Token]:
 997        """Returns a list of tokens corresponding to the SQL string `sql`."""
 998        if USE_RS_TOKENIZER:
 999            return self.tokenize_rs(sql)
1000
1001        self.reset()
1002        self.sql = sql
1003        self.size = len(sql)
1004
1005        try:
1006            self._scan()
1007        except Exception as e:
1008            start = max(self._current - 50, 0)
1009            end = min(self._current + 50, self.size - 1)
1010            context = self.sql[start:end]
1011            raise TokenError(f"Error tokenizing '{context}'") from e
1012
1013        return self.tokens
1014
1015    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
1016        while self.size and not self._end:
1017            current = self._current
1018
1019            # Skip spaces here rather than iteratively calling advance() for performance reasons
1020            while current < self.size:
1021                char = self.sql[current]
1022
1023                if char.isspace() and (char == " " or char == "\t"):
1024                    current += 1
1025                else:
1026                    break
1027
1028            offset = current - self._current if current > self._current else 1
1029
1030            self._start = current
1031            self._advance(offset)
1032
1033            if not self._char.isspace():
1034                if self._char.isdigit():
1035                    self._scan_number()
1036                elif self._char in self._IDENTIFIERS:
1037                    self._scan_identifier(self._IDENTIFIERS[self._char])
1038                else:
1039                    self._scan_keywords()
1040
1041            if until and until():
1042                break
1043
1044        if self.tokens and self._comments:
1045            self.tokens[-1].comments.extend(self._comments)
1046
1047    def _chars(self, size: int) -> str:
1048        if size == 1:
1049            return self._char
1050
1051        start = self._current - 1
1052        end = start + size
1053
1054        return self.sql[start:end] if end <= self.size else ""
1055
1056    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1057        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1058            # Ensures we don't count an extra line if we get a \r\n line break sequence
1059            if not (self._char == "\r" and self._peek == "\n"):
1060                self._col = 1
1061                self._line += 1
1062        else:
1063            self._col += i
1064
1065        self._current += i
1066        self._end = self._current >= self.size
1067        self._char = self.sql[self._current - 1]
1068        self._peek = "" if self._end else self.sql[self._current]
1069
1070        if alnum and self._char.isalnum():
1071            # Here we use local variables instead of attributes for better performance
1072            _col = self._col
1073            _current = self._current
1074            _end = self._end
1075            _peek = self._peek
1076
1077            while _peek.isalnum():
1078                _col += 1
1079                _current += 1
1080                _end = _current >= self.size
1081                _peek = "" if _end else self.sql[_current]
1082
1083            self._col = _col
1084            self._current = _current
1085            self._end = _end
1086            self._peek = _peek
1087            self._char = self.sql[_current - 1]
1088
1089    @property
1090    def _text(self) -> str:
1091        return self.sql[self._start : self._current]
1092
1093    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1094        self._prev_token_line = self._line
1095
1096        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1097            self.tokens[-1].comments.extend(self._comments)
1098            self._comments = []
1099
1100        self.tokens.append(
1101            Token(
1102                token_type,
1103                text=self._text if text is None else text,
1104                line=self._line,
1105                col=self._col,
1106                start=self._start,
1107                end=self._current - 1,
1108                comments=self._comments,
1109            )
1110        )
1111        self._comments = []
1112
1113        # If we have either a semicolon or a begin token before the command's token, we'll parse
1114        # whatever follows the command's token as a string
1115        if (
1116            token_type in self.COMMANDS
1117            and self._peek != ";"
1118            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1119        ):
1120            start = self._current
1121            tokens = len(self.tokens)
1122            self._scan(lambda: self._peek == ";")
1123            self.tokens = self.tokens[:tokens]
1124            text = self.sql[start : self._current].strip()
1125            if text:
1126                self._add(TokenType.STRING, text)
1127
1128    def _scan_keywords(self) -> None:
1129        size = 0
1130        word = None
1131        chars = self._text
1132        char = chars
1133        prev_space = False
1134        skip = False
1135        trie = self._KEYWORD_TRIE
1136        single_token = char in self.SINGLE_TOKENS
1137
1138        while chars:
1139            if skip:
1140                result = TrieResult.PREFIX
1141            else:
1142                result, trie = in_trie(trie, char.upper())
1143
1144            if result == TrieResult.FAILED:
1145                break
1146            if result == TrieResult.EXISTS:
1147                word = chars
1148
1149            end = self._current + size
1150            size += 1
1151
1152            if end < self.size:
1153                char = self.sql[end]
1154                single_token = single_token or char in self.SINGLE_TOKENS
1155                is_space = char.isspace()
1156
1157                if not is_space or not prev_space:
1158                    if is_space:
1159                        char = " "
1160                    chars += char
1161                    prev_space = is_space
1162                    skip = False
1163                else:
1164                    skip = True
1165            else:
1166                char = ""
1167                break
1168
1169        if word:
1170            if self._scan_string(word):
1171                return
1172            if self._scan_comment(word):
1173                return
1174            if prev_space or single_token or not char:
1175                self._advance(size - 1)
1176                word = word.upper()
1177                self._add(self.KEYWORDS[word], text=word)
1178                return
1179
1180        if self._char in self.SINGLE_TOKENS:
1181            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1182            return
1183
1184        self._scan_var()
1185
1186    def _scan_comment(self, comment_start: str) -> bool:
1187        if comment_start not in self._COMMENTS:
1188            return False
1189
1190        comment_start_line = self._line
1191        comment_start_size = len(comment_start)
1192        comment_end = self._COMMENTS[comment_start]
1193
1194        if comment_end:
1195            # Skip the comment's start delimiter
1196            self._advance(comment_start_size)
1197
1198            comment_count = 1
1199            comment_end_size = len(comment_end)
1200
1201            while not self._end:
1202                if self._chars(comment_end_size) == comment_end:
1203                    comment_count -= 1
1204                    if not comment_count:
1205                        break
1206
1207                self._advance(alnum=True)
1208
1209                # Nested comments are allowed by some dialects, e.g. databricks, duckdb, postgres
1210                if (
1211                    self.NESTED_COMMENTS
1212                    and not self._end
1213                    and self._chars(comment_end_size) == comment_start
1214                ):
1215                    self._advance(comment_start_size)
1216                    comment_count += 1
1217
1218            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1219            self._advance(comment_end_size - 1)
1220        else:
1221            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1222                self._advance(alnum=True)
1223            self._comments.append(self._text[comment_start_size:])
1224
1225        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1226        # Multiple consecutive comments are preserved by appending them to the current comments list.
1227        if comment_start_line == self._prev_token_line:
1228            self.tokens[-1].comments.extend(self._comments)
1229            self._comments = []
1230            self._prev_token_line = self._line
1231
1232        return True
1233
1234    def _scan_number(self) -> None:
1235        if self._char == "0":
1236            peek = self._peek.upper()
1237            if peek == "B":
1238                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1239            elif peek == "X":
1240                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1241
1242        decimal = False
1243        scientific = 0
1244
1245        while True:
1246            if self._peek.isdigit():
1247                self._advance()
1248            elif self._peek == "." and not decimal:
1249                if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER:
1250                    return self._add(TokenType.NUMBER)
1251                decimal = True
1252                self._advance()
1253            elif self._peek in ("-", "+") and scientific == 1:
1254                scientific += 1
1255                self._advance()
1256            elif self._peek.upper() == "E" and not scientific:
1257                scientific += 1
1258                self._advance()
1259            elif self._peek.isidentifier():
1260                number_text = self._text
1261                literal = ""
1262
1263                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1264                    literal += self._peek
1265                    self._advance()
1266
1267                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1268
1269                if token_type:
1270                    self._add(TokenType.NUMBER, number_text)
1271                    self._add(TokenType.DCOLON, "::")
1272                    return self._add(token_type, literal)
1273                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1274                    return self._add(TokenType.VAR)
1275
1276                self._advance(-len(literal))
1277                return self._add(TokenType.NUMBER, number_text)
1278            else:
1279                return self._add(TokenType.NUMBER)
1280
1281    def _scan_bits(self) -> None:
1282        self._advance()
1283        value = self._extract_value()
1284        try:
1285            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1286            int(value, 2)
1287            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1288        except ValueError:
1289            self._add(TokenType.IDENTIFIER)
1290
1291    def _scan_hex(self) -> None:
1292        self._advance()
1293        value = self._extract_value()
1294        try:
1295            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1296            int(value, 16)
1297            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1298        except ValueError:
1299            self._add(TokenType.IDENTIFIER)
1300
1301    def _extract_value(self) -> str:
1302        while True:
1303            char = self._peek.strip()
1304            if char and char not in self.SINGLE_TOKENS:
1305                self._advance(alnum=True)
1306            else:
1307                break
1308
1309        return self._text
1310
1311    def _scan_string(self, start: str) -> bool:
1312        base = None
1313        token_type = TokenType.STRING
1314
1315        if start in self._QUOTES:
1316            end = self._QUOTES[start]
1317        elif start in self._FORMAT_STRINGS:
1318            end, token_type = self._FORMAT_STRINGS[start]
1319
1320            if token_type == TokenType.HEX_STRING:
1321                base = 16
1322            elif token_type == TokenType.BIT_STRING:
1323                base = 2
1324            elif token_type == TokenType.HEREDOC_STRING:
1325                self._advance()
1326
1327                if self._char == end:
1328                    tag = ""
1329                else:
1330                    tag = self._extract_string(
1331                        end,
1332                        raw_string=True,
1333                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1334                    )
1335
1336                if tag and self.HEREDOC_TAG_IS_IDENTIFIER and (self._end or not tag.isidentifier()):
1337                    if not self._end:
1338                        self._advance(-1)
1339
1340                    self._advance(-len(tag))
1341                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1342                    return True
1343
1344                end = f"{start}{tag}{end}"
1345        else:
1346            return False
1347
1348        self._advance(len(start))
1349        text = self._extract_string(end, raw_string=token_type == TokenType.RAW_STRING)
1350
1351        if base:
1352            try:
1353                int(text, base)
1354            except Exception:
1355                raise TokenError(
1356                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1357                )
1358
1359        self._add(token_type, text)
1360        return True
1361
1362    def _scan_identifier(self, identifier_end: str) -> None:
1363        self._advance()
1364        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1365        self._add(TokenType.IDENTIFIER, text)
1366
1367    def _scan_var(self) -> None:
1368        while True:
1369            char = self._peek.strip()
1370            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1371                self._advance(alnum=True)
1372            else:
1373                break
1374
1375        self._add(
1376            TokenType.VAR
1377            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1378            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1379        )
1380
1381    def _extract_string(
1382        self,
1383        delimiter: str,
1384        escapes: t.Optional[t.Set[str]] = None,
1385        raw_string: bool = False,
1386        raise_unmatched: bool = True,
1387    ) -> str:
1388        text = ""
1389        delim_size = len(delimiter)
1390        escapes = self._STRING_ESCAPES if escapes is None else escapes
1391
1392        while True:
1393            if (
1394                not raw_string
1395                and self.dialect.UNESCAPED_SEQUENCES
1396                and self._peek
1397                and self._char in self.STRING_ESCAPES
1398            ):
1399                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1400                if unescaped_sequence:
1401                    self._advance(2)
1402                    text += unescaped_sequence
1403                    continue
1404            if (
1405                (self.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS or not raw_string)
1406                and self._char in escapes
1407                and (self._peek == delimiter or self._peek in escapes)
1408                and (self._char not in self._QUOTES or self._char == self._peek)
1409            ):
1410                if self._peek == delimiter:
1411                    text += self._peek
1412                else:
1413                    text += self._char + self._peek
1414
1415                if self._current + 1 < self.size:
1416                    self._advance(2)
1417                else:
1418                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1419            else:
1420                if self._chars(delim_size) == delimiter:
1421                    if delim_size > 1:
1422                        self._advance(delim_size - 1)
1423                    break
1424
1425                if self._end:
1426                    if not raise_unmatched:
1427                        return text + self._char
1428
1429                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1430
1431                current = self._current - 1
1432                self._advance(alnum=True)
1433                text += self.sql[current : self._current - 1]
1434
1435        return text
1436
1437    def tokenize_rs(self, sql: str) -> t.List[Token]:
1438        if not self._RS_TOKENIZER:
1439            raise SqlglotError("Rust tokenizer is not available")
1440
1441        try:
1442            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1443            for token in tokens:
1444                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1445            return tokens
1446        except Exception as e:
1447            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    CARET_AT = auto()
 65    TILDA = auto()
 66    ARROW = auto()
 67    DARROW = auto()
 68    FARROW = auto()
 69    HASH = auto()
 70    HASH_ARROW = auto()
 71    DHASH_ARROW = auto()
 72    LR_ARROW = auto()
 73    DAT = auto()
 74    LT_AT = auto()
 75    AT_GT = auto()
 76    DOLLAR = auto()
 77    PARAMETER = auto()
 78    SESSION_PARAMETER = auto()
 79    DAMP = auto()
 80    XOR = auto()
 81    DSTAR = auto()
 82
 83    BLOCK_START = auto()
 84    BLOCK_END = auto()
 85
 86    SPACE = auto()
 87    BREAK = auto()
 88
 89    STRING = auto()
 90    NUMBER = auto()
 91    IDENTIFIER = auto()
 92    DATABASE = auto()
 93    COLUMN = auto()
 94    COLUMN_DEF = auto()
 95    SCHEMA = auto()
 96    TABLE = auto()
 97    WAREHOUSE = auto()
 98    STREAMLIT = auto()
 99    VAR = auto()
100    BIT_STRING = auto()
101    HEX_STRING = auto()
102    BYTE_STRING = auto()
103    NATIONAL_STRING = auto()
104    RAW_STRING = auto()
105    HEREDOC_STRING = auto()
106    UNICODE_STRING = auto()
107
108    # types
109    BIT = auto()
110    BOOLEAN = auto()
111    TINYINT = auto()
112    UTINYINT = auto()
113    SMALLINT = auto()
114    USMALLINT = auto()
115    MEDIUMINT = auto()
116    UMEDIUMINT = auto()
117    INT = auto()
118    UINT = auto()
119    BIGINT = auto()
120    UBIGINT = auto()
121    INT128 = auto()
122    UINT128 = auto()
123    INT256 = auto()
124    UINT256 = auto()
125    FLOAT = auto()
126    DOUBLE = auto()
127    DECIMAL = auto()
128    DECIMAL32 = auto()
129    DECIMAL64 = auto()
130    DECIMAL128 = auto()
131    UDECIMAL = auto()
132    BIGDECIMAL = auto()
133    CHAR = auto()
134    NCHAR = auto()
135    VARCHAR = auto()
136    NVARCHAR = auto()
137    BPCHAR = auto()
138    TEXT = auto()
139    MEDIUMTEXT = auto()
140    LONGTEXT = auto()
141    MEDIUMBLOB = auto()
142    LONGBLOB = auto()
143    TINYBLOB = auto()
144    TINYTEXT = auto()
145    NAME = auto()
146    BINARY = auto()
147    VARBINARY = auto()
148    JSON = auto()
149    JSONB = auto()
150    TIME = auto()
151    TIMETZ = auto()
152    TIMESTAMP = auto()
153    TIMESTAMPTZ = auto()
154    TIMESTAMPLTZ = auto()
155    TIMESTAMPNTZ = auto()
156    TIMESTAMP_S = auto()
157    TIMESTAMP_MS = auto()
158    TIMESTAMP_NS = auto()
159    DATETIME = auto()
160    DATETIME64 = auto()
161    DATE = auto()
162    DATE32 = auto()
163    INT4RANGE = auto()
164    INT4MULTIRANGE = auto()
165    INT8RANGE = auto()
166    INT8MULTIRANGE = auto()
167    NUMRANGE = auto()
168    NUMMULTIRANGE = auto()
169    TSRANGE = auto()
170    TSMULTIRANGE = auto()
171    TSTZRANGE = auto()
172    TSTZMULTIRANGE = auto()
173    DATERANGE = auto()
174    DATEMULTIRANGE = auto()
175    UUID = auto()
176    GEOGRAPHY = auto()
177    NULLABLE = auto()
178    GEOMETRY = auto()
179    POINT = auto()
180    RING = auto()
181    LINESTRING = auto()
182    MULTILINESTRING = auto()
183    POLYGON = auto()
184    MULTIPOLYGON = auto()
185    HLLSKETCH = auto()
186    HSTORE = auto()
187    SUPER = auto()
188    SERIAL = auto()
189    SMALLSERIAL = auto()
190    BIGSERIAL = auto()
191    XML = auto()
192    YEAR = auto()
193    UNIQUEIDENTIFIER = auto()
194    USERDEFINED = auto()
195    MONEY = auto()
196    SMALLMONEY = auto()
197    ROWVERSION = auto()
198    IMAGE = auto()
199    VARIANT = auto()
200    OBJECT = auto()
201    INET = auto()
202    IPADDRESS = auto()
203    IPPREFIX = auto()
204    IPV4 = auto()
205    IPV6 = auto()
206    ENUM = auto()
207    ENUM8 = auto()
208    ENUM16 = auto()
209    FIXEDSTRING = auto()
210    LOWCARDINALITY = auto()
211    NESTED = auto()
212    AGGREGATEFUNCTION = auto()
213    SIMPLEAGGREGATEFUNCTION = auto()
214    TDIGEST = auto()
215    UNKNOWN = auto()
216    VECTOR = auto()
217
218    # keywords
219    ALIAS = auto()
220    ALTER = auto()
221    ALWAYS = auto()
222    ALL = auto()
223    ANTI = auto()
224    ANY = auto()
225    APPLY = auto()
226    ARRAY = auto()
227    ASC = auto()
228    ASOF = auto()
229    AUTO_INCREMENT = auto()
230    BEGIN = auto()
231    BETWEEN = auto()
232    BULK_COLLECT_INTO = auto()
233    CACHE = auto()
234    CASE = auto()
235    CHARACTER_SET = auto()
236    CLUSTER_BY = auto()
237    COLLATE = auto()
238    COMMAND = auto()
239    COMMENT = auto()
240    COMMIT = auto()
241    CONNECT_BY = auto()
242    CONSTRAINT = auto()
243    COPY = auto()
244    CREATE = auto()
245    CROSS = auto()
246    CUBE = auto()
247    CURRENT_DATE = auto()
248    CURRENT_DATETIME = auto()
249    CURRENT_TIME = auto()
250    CURRENT_TIMESTAMP = auto()
251    CURRENT_USER = auto()
252    DECLARE = auto()
253    DEFAULT = auto()
254    DELETE = auto()
255    DESC = auto()
256    DESCRIBE = auto()
257    DICTIONARY = auto()
258    DISTINCT = auto()
259    DISTRIBUTE_BY = auto()
260    DIV = auto()
261    DROP = auto()
262    ELSE = auto()
263    END = auto()
264    ESCAPE = auto()
265    EXCEPT = auto()
266    EXECUTE = auto()
267    EXISTS = auto()
268    FALSE = auto()
269    FETCH = auto()
270    FILTER = auto()
271    FINAL = auto()
272    FIRST = auto()
273    FOR = auto()
274    FORCE = auto()
275    FOREIGN_KEY = auto()
276    FORMAT = auto()
277    FROM = auto()
278    FULL = auto()
279    FUNCTION = auto()
280    GLOB = auto()
281    GLOBAL = auto()
282    GRANT = auto()
283    GROUP_BY = auto()
284    GROUPING_SETS = auto()
285    HAVING = auto()
286    HINT = auto()
287    IGNORE = auto()
288    ILIKE = auto()
289    ILIKE_ANY = auto()
290    IN = auto()
291    INDEX = auto()
292    INNER = auto()
293    INSERT = auto()
294    INTERSECT = auto()
295    INTERVAL = auto()
296    INTO = auto()
297    INTRODUCER = auto()
298    IRLIKE = auto()
299    IS = auto()
300    ISNULL = auto()
301    JOIN = auto()
302    JOIN_MARKER = auto()
303    KEEP = auto()
304    KEY = auto()
305    KILL = auto()
306    LANGUAGE = auto()
307    LATERAL = auto()
308    LEFT = auto()
309    LIKE = auto()
310    LIKE_ANY = auto()
311    LIMIT = auto()
312    LIST = auto()
313    LOAD = auto()
314    LOCK = auto()
315    MAP = auto()
316    MATCH_CONDITION = auto()
317    MATCH_RECOGNIZE = auto()
318    MEMBER_OF = auto()
319    MERGE = auto()
320    MOD = auto()
321    MODEL = auto()
322    NATURAL = auto()
323    NEXT = auto()
324    NOTNULL = auto()
325    NULL = auto()
326    OBJECT_IDENTIFIER = auto()
327    OFFSET = auto()
328    ON = auto()
329    ONLY = auto()
330    OPERATOR = auto()
331    ORDER_BY = auto()
332    ORDER_SIBLINGS_BY = auto()
333    ORDERED = auto()
334    ORDINALITY = auto()
335    OUTER = auto()
336    OVER = auto()
337    OVERLAPS = auto()
338    OVERWRITE = auto()
339    PARTITION = auto()
340    PARTITION_BY = auto()
341    PERCENT = auto()
342    PIVOT = auto()
343    PLACEHOLDER = auto()
344    POSITIONAL = auto()
345    PRAGMA = auto()
346    PREWHERE = auto()
347    PRIMARY_KEY = auto()
348    PROCEDURE = auto()
349    PROPERTIES = auto()
350    PSEUDO_TYPE = auto()
351    QUALIFY = auto()
352    QUOTE = auto()
353    RANGE = auto()
354    RECURSIVE = auto()
355    REFRESH = auto()
356    RENAME = auto()
357    REPLACE = auto()
358    RETURNING = auto()
359    REFERENCES = auto()
360    RIGHT = auto()
361    RLIKE = auto()
362    ROLLBACK = auto()
363    ROLLUP = auto()
364    ROW = auto()
365    ROWS = auto()
366    SELECT = auto()
367    SEMI = auto()
368    SEPARATOR = auto()
369    SEQUENCE = auto()
370    SERDE_PROPERTIES = auto()
371    SET = auto()
372    SETTINGS = auto()
373    SHOW = auto()
374    SIMILAR_TO = auto()
375    SOME = auto()
376    SORT_BY = auto()
377    START_WITH = auto()
378    STORAGE_INTEGRATION = auto()
379    STRAIGHT_JOIN = auto()
380    STRUCT = auto()
381    SUMMARIZE = auto()
382    TABLE_SAMPLE = auto()
383    TAG = auto()
384    TEMPORARY = auto()
385    TOP = auto()
386    THEN = auto()
387    TRUE = auto()
388    TRUNCATE = auto()
389    UNCACHE = auto()
390    UNION = auto()
391    UNNEST = auto()
392    UNPIVOT = auto()
393    UPDATE = auto()
394    USE = auto()
395    USING = auto()
396    VALUES = auto()
397    VIEW = auto()
398    VOLATILE = auto()
399    WHEN = auto()
400    WHERE = auto()
401    WINDOW = auto()
402    WITH = auto()
403    UNIQUE = auto()
404    VERSION_SNAPSHOT = auto()
405    TIMESTAMP_SNAPSHOT = auto()
406    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'>
CARET_AT = <TokenType.CARET_AT: 'CARET_AT'>
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'>
POINT = <TokenType.POINT: 'POINT'>
RING = <TokenType.RING: 'RING'>
LINESTRING = <TokenType.LINESTRING: 'LINESTRING'>
MULTILINESTRING = <TokenType.MULTILINESTRING: 'MULTILINESTRING'>
POLYGON = <TokenType.POLYGON: 'POLYGON'>
MULTIPOLYGON = <TokenType.MULTIPOLYGON: 'MULTIPOLYGON'>
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'>
class Token:
413class Token:
414    __slots__ = ("token_type", "text", "line", "col", "start", "end", "comments")
415
416    @classmethod
417    def number(cls, number: int) -> Token:
418        """Returns a NUMBER token with `number` as its text."""
419        return cls(TokenType.NUMBER, str(number))
420
421    @classmethod
422    def string(cls, string: str) -> Token:
423        """Returns a STRING token with `string` as its text."""
424        return cls(TokenType.STRING, string)
425
426    @classmethod
427    def identifier(cls, identifier: str) -> Token:
428        """Returns an IDENTIFIER token with `identifier` as its text."""
429        return cls(TokenType.IDENTIFIER, identifier)
430
431    @classmethod
432    def var(cls, var: str) -> Token:
433        """Returns an VAR token with `var` as its text."""
434        return cls(TokenType.VAR, var)
435
436    def __init__(
437        self,
438        token_type: TokenType,
439        text: str,
440        line: int = 1,
441        col: int = 1,
442        start: int = 0,
443        end: int = 0,
444        comments: t.Optional[t.List[str]] = None,
445    ) -> None:
446        """Token initializer.
447
448        Args:
449            token_type: The TokenType Enum.
450            text: The text of the token.
451            line: The line that the token ends on.
452            col: The column that the token ends on.
453            start: The start index of the token.
454            end: The ending index of the token.
455            comments: The comments to attach to the token.
456        """
457        self.token_type = token_type
458        self.text = text
459        self.line = line
460        self.col = col
461        self.start = start
462        self.end = end
463        self.comments = [] if comments is None else comments
464
465    def __repr__(self) -> str:
466        attributes = ", ".join(f"{k}: {getattr(self, k)}" for k in self.__slots__)
467        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)
436    def __init__(
437        self,
438        token_type: TokenType,
439        text: str,
440        line: int = 1,
441        col: int = 1,
442        start: int = 0,
443        end: int = 0,
444        comments: t.Optional[t.List[str]] = None,
445    ) -> None:
446        """Token initializer.
447
448        Args:
449            token_type: The TokenType Enum.
450            text: The text of the token.
451            line: The line that the token ends on.
452            col: The column that the token ends on.
453            start: The start index of the token.
454            end: The ending index of the token.
455            comments: The comments to attach to the token.
456        """
457        self.token_type = token_type
458        self.text = text
459        self.line = line
460        self.col = col
461        self.start = start
462        self.end = end
463        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:
416    @classmethod
417    def number(cls, number: int) -> Token:
418        """Returns a NUMBER token with `number` as its text."""
419        return cls(TokenType.NUMBER, str(number))

Returns a NUMBER token with number as its text.

@classmethod
def string(cls, string: str) -> Token:
421    @classmethod
422    def string(cls, string: str) -> Token:
423        """Returns a STRING token with `string` as its text."""
424        return cls(TokenType.STRING, string)

Returns a STRING token with string as its text.

@classmethod
def identifier(cls, identifier: str) -> Token:
426    @classmethod
427    def identifier(cls, identifier: str) -> Token:
428        """Returns an IDENTIFIER token with `identifier` as its text."""
429        return cls(TokenType.IDENTIFIER, identifier)

Returns an IDENTIFIER token with identifier as its text.

@classmethod
def var(cls, var: str) -> Token:
431    @classmethod
432    def var(cls, var: str) -> Token:
433        """Returns an VAR token with `var` as its text."""
434        return cls(TokenType.VAR, var)

Returns an VAR token with var as its text.

token_type
text
line
col
start
end
comments
class Tokenizer:
 570class Tokenizer(metaclass=_Tokenizer):
 571    SINGLE_TOKENS = {
 572        "(": TokenType.L_PAREN,
 573        ")": TokenType.R_PAREN,
 574        "[": TokenType.L_BRACKET,
 575        "]": TokenType.R_BRACKET,
 576        "{": TokenType.L_BRACE,
 577        "}": TokenType.R_BRACE,
 578        "&": TokenType.AMP,
 579        "^": TokenType.CARET,
 580        ":": TokenType.COLON,
 581        ",": TokenType.COMMA,
 582        ".": TokenType.DOT,
 583        "-": TokenType.DASH,
 584        "=": TokenType.EQ,
 585        ">": TokenType.GT,
 586        "<": TokenType.LT,
 587        "%": TokenType.MOD,
 588        "!": TokenType.NOT,
 589        "|": TokenType.PIPE,
 590        "+": TokenType.PLUS,
 591        ";": TokenType.SEMICOLON,
 592        "/": TokenType.SLASH,
 593        "\\": TokenType.BACKSLASH,
 594        "*": TokenType.STAR,
 595        "~": TokenType.TILDA,
 596        "?": TokenType.PLACEHOLDER,
 597        "@": TokenType.PARAMETER,
 598        "#": TokenType.HASH,
 599        # Used for breaking a var like x'y' but nothing else the token type doesn't matter
 600        "'": TokenType.UNKNOWN,
 601        "`": TokenType.UNKNOWN,
 602        '"': TokenType.UNKNOWN,
 603    }
 604
 605    BIT_STRINGS: t.List[str | t.Tuple[str, str]] = []
 606    BYTE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 607    HEX_STRINGS: t.List[str | t.Tuple[str, str]] = []
 608    RAW_STRINGS: t.List[str | t.Tuple[str, str]] = []
 609    HEREDOC_STRINGS: t.List[str | t.Tuple[str, str]] = []
 610    UNICODE_STRINGS: t.List[str | t.Tuple[str, str]] = []
 611    IDENTIFIERS: t.List[str | t.Tuple[str, str]] = ['"']
 612    IDENTIFIER_ESCAPES = ['"']
 613    QUOTES: t.List[t.Tuple[str, str] | str] = ["'"]
 614    STRING_ESCAPES = ["'"]
 615    VAR_SINGLE_TOKENS: t.Set[str] = set()
 616
 617    # Whether the heredoc tags follow the same lexical rules as unquoted identifiers
 618    HEREDOC_TAG_IS_IDENTIFIER = False
 619
 620    # Token that we'll generate as a fallback if the heredoc prefix doesn't correspond to a heredoc
 621    HEREDOC_STRING_ALTERNATIVE = TokenType.VAR
 622
 623    # Whether string escape characters function as such when placed within raw strings
 624    STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS = True
 625
 626    NESTED_COMMENTS = True
 627
 628    # Autofilled
 629    _COMMENTS: t.Dict[str, str] = {}
 630    _FORMAT_STRINGS: t.Dict[str, t.Tuple[str, TokenType]] = {}
 631    _IDENTIFIERS: t.Dict[str, str] = {}
 632    _IDENTIFIER_ESCAPES: t.Set[str] = set()
 633    _QUOTES: t.Dict[str, str] = {}
 634    _STRING_ESCAPES: t.Set[str] = set()
 635    _KEYWORD_TRIE: t.Dict = {}
 636    _RS_TOKENIZER: t.Optional[t.Any] = None
 637
 638    KEYWORDS: t.Dict[str, TokenType] = {
 639        **{f"{{%{postfix}": TokenType.BLOCK_START for postfix in ("", "+", "-")},
 640        **{f"{prefix}%}}": TokenType.BLOCK_END for prefix in ("", "+", "-")},
 641        **{f"{{{{{postfix}": TokenType.BLOCK_START for postfix in ("+", "-")},
 642        **{f"{prefix}}}}}": TokenType.BLOCK_END for prefix in ("+", "-")},
 643        "/*+": TokenType.HINT,
 644        "==": TokenType.EQ,
 645        "::": TokenType.DCOLON,
 646        "||": TokenType.DPIPE,
 647        ">=": TokenType.GTE,
 648        "<=": TokenType.LTE,
 649        "<>": TokenType.NEQ,
 650        "!=": TokenType.NEQ,
 651        ":=": TokenType.COLON_EQ,
 652        "<=>": TokenType.NULLSAFE_EQ,
 653        "->": TokenType.ARROW,
 654        "->>": TokenType.DARROW,
 655        "=>": TokenType.FARROW,
 656        "#>": TokenType.HASH_ARROW,
 657        "#>>": TokenType.DHASH_ARROW,
 658        "<->": TokenType.LR_ARROW,
 659        "&&": TokenType.DAMP,
 660        "??": TokenType.DQMARK,
 661        "~~~": TokenType.GLOB,
 662        "~~": TokenType.LIKE,
 663        "~~*": TokenType.ILIKE,
 664        "~*": TokenType.IRLIKE,
 665        "ALL": TokenType.ALL,
 666        "ALWAYS": TokenType.ALWAYS,
 667        "AND": TokenType.AND,
 668        "ANTI": TokenType.ANTI,
 669        "ANY": TokenType.ANY,
 670        "ASC": TokenType.ASC,
 671        "AS": TokenType.ALIAS,
 672        "ASOF": TokenType.ASOF,
 673        "AUTOINCREMENT": TokenType.AUTO_INCREMENT,
 674        "AUTO_INCREMENT": TokenType.AUTO_INCREMENT,
 675        "BEGIN": TokenType.BEGIN,
 676        "BETWEEN": TokenType.BETWEEN,
 677        "CACHE": TokenType.CACHE,
 678        "UNCACHE": TokenType.UNCACHE,
 679        "CASE": TokenType.CASE,
 680        "CHARACTER SET": TokenType.CHARACTER_SET,
 681        "CLUSTER BY": TokenType.CLUSTER_BY,
 682        "COLLATE": TokenType.COLLATE,
 683        "COLUMN": TokenType.COLUMN,
 684        "COMMIT": TokenType.COMMIT,
 685        "CONNECT BY": TokenType.CONNECT_BY,
 686        "CONSTRAINT": TokenType.CONSTRAINT,
 687        "COPY": TokenType.COPY,
 688        "CREATE": TokenType.CREATE,
 689        "CROSS": TokenType.CROSS,
 690        "CUBE": TokenType.CUBE,
 691        "CURRENT_DATE": TokenType.CURRENT_DATE,
 692        "CURRENT_TIME": TokenType.CURRENT_TIME,
 693        "CURRENT_TIMESTAMP": TokenType.CURRENT_TIMESTAMP,
 694        "CURRENT_USER": TokenType.CURRENT_USER,
 695        "DATABASE": TokenType.DATABASE,
 696        "DEFAULT": TokenType.DEFAULT,
 697        "DELETE": TokenType.DELETE,
 698        "DESC": TokenType.DESC,
 699        "DESCRIBE": TokenType.DESCRIBE,
 700        "DISTINCT": TokenType.DISTINCT,
 701        "DISTRIBUTE BY": TokenType.DISTRIBUTE_BY,
 702        "DIV": TokenType.DIV,
 703        "DROP": TokenType.DROP,
 704        "ELSE": TokenType.ELSE,
 705        "END": TokenType.END,
 706        "ENUM": TokenType.ENUM,
 707        "ESCAPE": TokenType.ESCAPE,
 708        "EXCEPT": TokenType.EXCEPT,
 709        "EXECUTE": TokenType.EXECUTE,
 710        "EXISTS": TokenType.EXISTS,
 711        "FALSE": TokenType.FALSE,
 712        "FETCH": TokenType.FETCH,
 713        "FILTER": TokenType.FILTER,
 714        "FIRST": TokenType.FIRST,
 715        "FULL": TokenType.FULL,
 716        "FUNCTION": TokenType.FUNCTION,
 717        "FOR": TokenType.FOR,
 718        "FOREIGN KEY": TokenType.FOREIGN_KEY,
 719        "FORMAT": TokenType.FORMAT,
 720        "FROM": TokenType.FROM,
 721        "GEOGRAPHY": TokenType.GEOGRAPHY,
 722        "GEOMETRY": TokenType.GEOMETRY,
 723        "GLOB": TokenType.GLOB,
 724        "GROUP BY": TokenType.GROUP_BY,
 725        "GROUPING SETS": TokenType.GROUPING_SETS,
 726        "HAVING": TokenType.HAVING,
 727        "ILIKE": TokenType.ILIKE,
 728        "IN": TokenType.IN,
 729        "INDEX": TokenType.INDEX,
 730        "INET": TokenType.INET,
 731        "INNER": TokenType.INNER,
 732        "INSERT": TokenType.INSERT,
 733        "INTERVAL": TokenType.INTERVAL,
 734        "INTERSECT": TokenType.INTERSECT,
 735        "INTO": TokenType.INTO,
 736        "IS": TokenType.IS,
 737        "ISNULL": TokenType.ISNULL,
 738        "JOIN": TokenType.JOIN,
 739        "KEEP": TokenType.KEEP,
 740        "KILL": TokenType.KILL,
 741        "LATERAL": TokenType.LATERAL,
 742        "LEFT": TokenType.LEFT,
 743        "LIKE": TokenType.LIKE,
 744        "LIMIT": TokenType.LIMIT,
 745        "LOAD": TokenType.LOAD,
 746        "LOCK": TokenType.LOCK,
 747        "MERGE": TokenType.MERGE,
 748        "NATURAL": TokenType.NATURAL,
 749        "NEXT": TokenType.NEXT,
 750        "NOT": TokenType.NOT,
 751        "NOTNULL": TokenType.NOTNULL,
 752        "NULL": TokenType.NULL,
 753        "OBJECT": TokenType.OBJECT,
 754        "OFFSET": TokenType.OFFSET,
 755        "ON": TokenType.ON,
 756        "OR": TokenType.OR,
 757        "XOR": TokenType.XOR,
 758        "ORDER BY": TokenType.ORDER_BY,
 759        "ORDINALITY": TokenType.ORDINALITY,
 760        "OUTER": TokenType.OUTER,
 761        "OVER": TokenType.OVER,
 762        "OVERLAPS": TokenType.OVERLAPS,
 763        "OVERWRITE": TokenType.OVERWRITE,
 764        "PARTITION": TokenType.PARTITION,
 765        "PARTITION BY": TokenType.PARTITION_BY,
 766        "PARTITIONED BY": TokenType.PARTITION_BY,
 767        "PARTITIONED_BY": TokenType.PARTITION_BY,
 768        "PERCENT": TokenType.PERCENT,
 769        "PIVOT": TokenType.PIVOT,
 770        "PRAGMA": TokenType.PRAGMA,
 771        "PRIMARY KEY": TokenType.PRIMARY_KEY,
 772        "PROCEDURE": TokenType.PROCEDURE,
 773        "QUALIFY": TokenType.QUALIFY,
 774        "RANGE": TokenType.RANGE,
 775        "RECURSIVE": TokenType.RECURSIVE,
 776        "REGEXP": TokenType.RLIKE,
 777        "RENAME": TokenType.RENAME,
 778        "REPLACE": TokenType.REPLACE,
 779        "RETURNING": TokenType.RETURNING,
 780        "REFERENCES": TokenType.REFERENCES,
 781        "RIGHT": TokenType.RIGHT,
 782        "RLIKE": TokenType.RLIKE,
 783        "ROLLBACK": TokenType.ROLLBACK,
 784        "ROLLUP": TokenType.ROLLUP,
 785        "ROW": TokenType.ROW,
 786        "ROWS": TokenType.ROWS,
 787        "SCHEMA": TokenType.SCHEMA,
 788        "SELECT": TokenType.SELECT,
 789        "SEMI": TokenType.SEMI,
 790        "SET": TokenType.SET,
 791        "SETTINGS": TokenType.SETTINGS,
 792        "SHOW": TokenType.SHOW,
 793        "SIMILAR TO": TokenType.SIMILAR_TO,
 794        "SOME": TokenType.SOME,
 795        "SORT BY": TokenType.SORT_BY,
 796        "START WITH": TokenType.START_WITH,
 797        "STRAIGHT_JOIN": TokenType.STRAIGHT_JOIN,
 798        "TABLE": TokenType.TABLE,
 799        "TABLESAMPLE": TokenType.TABLE_SAMPLE,
 800        "TEMP": TokenType.TEMPORARY,
 801        "TEMPORARY": TokenType.TEMPORARY,
 802        "THEN": TokenType.THEN,
 803        "TRUE": TokenType.TRUE,
 804        "TRUNCATE": TokenType.TRUNCATE,
 805        "UNION": TokenType.UNION,
 806        "UNKNOWN": TokenType.UNKNOWN,
 807        "UNNEST": TokenType.UNNEST,
 808        "UNPIVOT": TokenType.UNPIVOT,
 809        "UPDATE": TokenType.UPDATE,
 810        "USE": TokenType.USE,
 811        "USING": TokenType.USING,
 812        "UUID": TokenType.UUID,
 813        "VALUES": TokenType.VALUES,
 814        "VIEW": TokenType.VIEW,
 815        "VOLATILE": TokenType.VOLATILE,
 816        "WHEN": TokenType.WHEN,
 817        "WHERE": TokenType.WHERE,
 818        "WINDOW": TokenType.WINDOW,
 819        "WITH": TokenType.WITH,
 820        "APPLY": TokenType.APPLY,
 821        "ARRAY": TokenType.ARRAY,
 822        "BIT": TokenType.BIT,
 823        "BOOL": TokenType.BOOLEAN,
 824        "BOOLEAN": TokenType.BOOLEAN,
 825        "BYTE": TokenType.TINYINT,
 826        "MEDIUMINT": TokenType.MEDIUMINT,
 827        "INT1": TokenType.TINYINT,
 828        "TINYINT": TokenType.TINYINT,
 829        "INT16": TokenType.SMALLINT,
 830        "SHORT": TokenType.SMALLINT,
 831        "SMALLINT": TokenType.SMALLINT,
 832        "INT128": TokenType.INT128,
 833        "HUGEINT": TokenType.INT128,
 834        "UHUGEINT": TokenType.UINT128,
 835        "INT2": TokenType.SMALLINT,
 836        "INTEGER": TokenType.INT,
 837        "INT": TokenType.INT,
 838        "INT4": TokenType.INT,
 839        "INT32": TokenType.INT,
 840        "INT64": TokenType.BIGINT,
 841        "LONG": TokenType.BIGINT,
 842        "BIGINT": TokenType.BIGINT,
 843        "INT8": TokenType.TINYINT,
 844        "UINT": TokenType.UINT,
 845        "DEC": TokenType.DECIMAL,
 846        "DECIMAL": TokenType.DECIMAL,
 847        "DECIMAL32": TokenType.DECIMAL32,
 848        "DECIMAL64": TokenType.DECIMAL64,
 849        "DECIMAL128": TokenType.DECIMAL128,
 850        "BIGDECIMAL": TokenType.BIGDECIMAL,
 851        "BIGNUMERIC": TokenType.BIGDECIMAL,
 852        "LIST": TokenType.LIST,
 853        "MAP": TokenType.MAP,
 854        "NULLABLE": TokenType.NULLABLE,
 855        "NUMBER": TokenType.DECIMAL,
 856        "NUMERIC": TokenType.DECIMAL,
 857        "FIXED": TokenType.DECIMAL,
 858        "REAL": TokenType.FLOAT,
 859        "FLOAT": TokenType.FLOAT,
 860        "FLOAT4": TokenType.FLOAT,
 861        "FLOAT8": TokenType.DOUBLE,
 862        "DOUBLE": TokenType.DOUBLE,
 863        "DOUBLE PRECISION": TokenType.DOUBLE,
 864        "JSON": TokenType.JSON,
 865        "JSONB": TokenType.JSONB,
 866        "CHAR": TokenType.CHAR,
 867        "CHARACTER": TokenType.CHAR,
 868        "NCHAR": TokenType.NCHAR,
 869        "VARCHAR": TokenType.VARCHAR,
 870        "VARCHAR2": TokenType.VARCHAR,
 871        "NVARCHAR": TokenType.NVARCHAR,
 872        "NVARCHAR2": TokenType.NVARCHAR,
 873        "BPCHAR": TokenType.BPCHAR,
 874        "STR": TokenType.TEXT,
 875        "STRING": TokenType.TEXT,
 876        "TEXT": TokenType.TEXT,
 877        "LONGTEXT": TokenType.LONGTEXT,
 878        "MEDIUMTEXT": TokenType.MEDIUMTEXT,
 879        "TINYTEXT": TokenType.TINYTEXT,
 880        "CLOB": TokenType.TEXT,
 881        "LONGVARCHAR": TokenType.TEXT,
 882        "BINARY": TokenType.BINARY,
 883        "BLOB": TokenType.VARBINARY,
 884        "LONGBLOB": TokenType.LONGBLOB,
 885        "MEDIUMBLOB": TokenType.MEDIUMBLOB,
 886        "TINYBLOB": TokenType.TINYBLOB,
 887        "BYTEA": TokenType.VARBINARY,
 888        "VARBINARY": TokenType.VARBINARY,
 889        "TIME": TokenType.TIME,
 890        "TIMETZ": TokenType.TIMETZ,
 891        "TIMESTAMP": TokenType.TIMESTAMP,
 892        "TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
 893        "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
 894        "TIMESTAMP_LTZ": TokenType.TIMESTAMPLTZ,
 895        "TIMESTAMPNTZ": TokenType.TIMESTAMPNTZ,
 896        "TIMESTAMP_NTZ": TokenType.TIMESTAMPNTZ,
 897        "DATE": TokenType.DATE,
 898        "DATETIME": TokenType.DATETIME,
 899        "INT4RANGE": TokenType.INT4RANGE,
 900        "INT4MULTIRANGE": TokenType.INT4MULTIRANGE,
 901        "INT8RANGE": TokenType.INT8RANGE,
 902        "INT8MULTIRANGE": TokenType.INT8MULTIRANGE,
 903        "NUMRANGE": TokenType.NUMRANGE,
 904        "NUMMULTIRANGE": TokenType.NUMMULTIRANGE,
 905        "TSRANGE": TokenType.TSRANGE,
 906        "TSMULTIRANGE": TokenType.TSMULTIRANGE,
 907        "TSTZRANGE": TokenType.TSTZRANGE,
 908        "TSTZMULTIRANGE": TokenType.TSTZMULTIRANGE,
 909        "DATERANGE": TokenType.DATERANGE,
 910        "DATEMULTIRANGE": TokenType.DATEMULTIRANGE,
 911        "UNIQUE": TokenType.UNIQUE,
 912        "VECTOR": TokenType.VECTOR,
 913        "STRUCT": TokenType.STRUCT,
 914        "SEQUENCE": TokenType.SEQUENCE,
 915        "VARIANT": TokenType.VARIANT,
 916        "ALTER": TokenType.ALTER,
 917        "ANALYZE": TokenType.COMMAND,
 918        "CALL": TokenType.COMMAND,
 919        "COMMENT": TokenType.COMMENT,
 920        "EXPLAIN": TokenType.COMMAND,
 921        "GRANT": TokenType.GRANT,
 922        "OPTIMIZE": TokenType.COMMAND,
 923        "PREPARE": TokenType.COMMAND,
 924        "VACUUM": TokenType.COMMAND,
 925        "USER-DEFINED": TokenType.USERDEFINED,
 926        "FOR VERSION": TokenType.VERSION_SNAPSHOT,
 927        "FOR TIMESTAMP": TokenType.TIMESTAMP_SNAPSHOT,
 928    }
 929
 930    WHITE_SPACE: t.Dict[t.Optional[str], TokenType] = {
 931        " ": TokenType.SPACE,
 932        "\t": TokenType.SPACE,
 933        "\n": TokenType.BREAK,
 934        "\r": TokenType.BREAK,
 935    }
 936
 937    COMMANDS = {
 938        TokenType.COMMAND,
 939        TokenType.EXECUTE,
 940        TokenType.FETCH,
 941        TokenType.SHOW,
 942        TokenType.RENAME,
 943    }
 944
 945    COMMAND_PREFIX_TOKENS = {TokenType.SEMICOLON, TokenType.BEGIN}
 946
 947    # Handle numeric literals like in hive (3L = BIGINT)
 948    NUMERIC_LITERALS: t.Dict[str, str] = {}
 949
 950    COMMENTS = ["--", ("/*", "*/")]
 951
 952    __slots__ = (
 953        "sql",
 954        "size",
 955        "tokens",
 956        "dialect",
 957        "_start",
 958        "_current",
 959        "_line",
 960        "_col",
 961        "_comments",
 962        "_char",
 963        "_end",
 964        "_peek",
 965        "_prev_token_line",
 966        "_rs_dialect_settings",
 967    )
 968
 969    def __init__(self, dialect: DialectType = None) -> None:
 970        from sqlglot.dialects import Dialect
 971
 972        self.dialect = Dialect.get_or_raise(dialect)
 973
 974        if USE_RS_TOKENIZER:
 975            self._rs_dialect_settings = RsTokenizerDialectSettings(
 976                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
 977                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
 978            )
 979
 980        self.reset()
 981
 982    def reset(self) -> None:
 983        self.sql = ""
 984        self.size = 0
 985        self.tokens: t.List[Token] = []
 986        self._start = 0
 987        self._current = 0
 988        self._line = 1
 989        self._col = 0
 990        self._comments: t.List[str] = []
 991
 992        self._char = ""
 993        self._end = False
 994        self._peek = ""
 995        self._prev_token_line = -1
 996
 997    def tokenize(self, sql: str) -> t.List[Token]:
 998        """Returns a list of tokens corresponding to the SQL string `sql`."""
 999        if USE_RS_TOKENIZER:
1000            return self.tokenize_rs(sql)
1001
1002        self.reset()
1003        self.sql = sql
1004        self.size = len(sql)
1005
1006        try:
1007            self._scan()
1008        except Exception as e:
1009            start = max(self._current - 50, 0)
1010            end = min(self._current + 50, self.size - 1)
1011            context = self.sql[start:end]
1012            raise TokenError(f"Error tokenizing '{context}'") from e
1013
1014        return self.tokens
1015
1016    def _scan(self, until: t.Optional[t.Callable] = None) -> None:
1017        while self.size and not self._end:
1018            current = self._current
1019
1020            # Skip spaces here rather than iteratively calling advance() for performance reasons
1021            while current < self.size:
1022                char = self.sql[current]
1023
1024                if char.isspace() and (char == " " or char == "\t"):
1025                    current += 1
1026                else:
1027                    break
1028
1029            offset = current - self._current if current > self._current else 1
1030
1031            self._start = current
1032            self._advance(offset)
1033
1034            if not self._char.isspace():
1035                if self._char.isdigit():
1036                    self._scan_number()
1037                elif self._char in self._IDENTIFIERS:
1038                    self._scan_identifier(self._IDENTIFIERS[self._char])
1039                else:
1040                    self._scan_keywords()
1041
1042            if until and until():
1043                break
1044
1045        if self.tokens and self._comments:
1046            self.tokens[-1].comments.extend(self._comments)
1047
1048    def _chars(self, size: int) -> str:
1049        if size == 1:
1050            return self._char
1051
1052        start = self._current - 1
1053        end = start + size
1054
1055        return self.sql[start:end] if end <= self.size else ""
1056
1057    def _advance(self, i: int = 1, alnum: bool = False) -> None:
1058        if self.WHITE_SPACE.get(self._char) is TokenType.BREAK:
1059            # Ensures we don't count an extra line if we get a \r\n line break sequence
1060            if not (self._char == "\r" and self._peek == "\n"):
1061                self._col = 1
1062                self._line += 1
1063        else:
1064            self._col += i
1065
1066        self._current += i
1067        self._end = self._current >= self.size
1068        self._char = self.sql[self._current - 1]
1069        self._peek = "" if self._end else self.sql[self._current]
1070
1071        if alnum and self._char.isalnum():
1072            # Here we use local variables instead of attributes for better performance
1073            _col = self._col
1074            _current = self._current
1075            _end = self._end
1076            _peek = self._peek
1077
1078            while _peek.isalnum():
1079                _col += 1
1080                _current += 1
1081                _end = _current >= self.size
1082                _peek = "" if _end else self.sql[_current]
1083
1084            self._col = _col
1085            self._current = _current
1086            self._end = _end
1087            self._peek = _peek
1088            self._char = self.sql[_current - 1]
1089
1090    @property
1091    def _text(self) -> str:
1092        return self.sql[self._start : self._current]
1093
1094    def _add(self, token_type: TokenType, text: t.Optional[str] = None) -> None:
1095        self._prev_token_line = self._line
1096
1097        if self._comments and token_type == TokenType.SEMICOLON and self.tokens:
1098            self.tokens[-1].comments.extend(self._comments)
1099            self._comments = []
1100
1101        self.tokens.append(
1102            Token(
1103                token_type,
1104                text=self._text if text is None else text,
1105                line=self._line,
1106                col=self._col,
1107                start=self._start,
1108                end=self._current - 1,
1109                comments=self._comments,
1110            )
1111        )
1112        self._comments = []
1113
1114        # If we have either a semicolon or a begin token before the command's token, we'll parse
1115        # whatever follows the command's token as a string
1116        if (
1117            token_type in self.COMMANDS
1118            and self._peek != ";"
1119            and (len(self.tokens) == 1 or self.tokens[-2].token_type in self.COMMAND_PREFIX_TOKENS)
1120        ):
1121            start = self._current
1122            tokens = len(self.tokens)
1123            self._scan(lambda: self._peek == ";")
1124            self.tokens = self.tokens[:tokens]
1125            text = self.sql[start : self._current].strip()
1126            if text:
1127                self._add(TokenType.STRING, text)
1128
1129    def _scan_keywords(self) -> None:
1130        size = 0
1131        word = None
1132        chars = self._text
1133        char = chars
1134        prev_space = False
1135        skip = False
1136        trie = self._KEYWORD_TRIE
1137        single_token = char in self.SINGLE_TOKENS
1138
1139        while chars:
1140            if skip:
1141                result = TrieResult.PREFIX
1142            else:
1143                result, trie = in_trie(trie, char.upper())
1144
1145            if result == TrieResult.FAILED:
1146                break
1147            if result == TrieResult.EXISTS:
1148                word = chars
1149
1150            end = self._current + size
1151            size += 1
1152
1153            if end < self.size:
1154                char = self.sql[end]
1155                single_token = single_token or char in self.SINGLE_TOKENS
1156                is_space = char.isspace()
1157
1158                if not is_space or not prev_space:
1159                    if is_space:
1160                        char = " "
1161                    chars += char
1162                    prev_space = is_space
1163                    skip = False
1164                else:
1165                    skip = True
1166            else:
1167                char = ""
1168                break
1169
1170        if word:
1171            if self._scan_string(word):
1172                return
1173            if self._scan_comment(word):
1174                return
1175            if prev_space or single_token or not char:
1176                self._advance(size - 1)
1177                word = word.upper()
1178                self._add(self.KEYWORDS[word], text=word)
1179                return
1180
1181        if self._char in self.SINGLE_TOKENS:
1182            self._add(self.SINGLE_TOKENS[self._char], text=self._char)
1183            return
1184
1185        self._scan_var()
1186
1187    def _scan_comment(self, comment_start: str) -> bool:
1188        if comment_start not in self._COMMENTS:
1189            return False
1190
1191        comment_start_line = self._line
1192        comment_start_size = len(comment_start)
1193        comment_end = self._COMMENTS[comment_start]
1194
1195        if comment_end:
1196            # Skip the comment's start delimiter
1197            self._advance(comment_start_size)
1198
1199            comment_count = 1
1200            comment_end_size = len(comment_end)
1201
1202            while not self._end:
1203                if self._chars(comment_end_size) == comment_end:
1204                    comment_count -= 1
1205                    if not comment_count:
1206                        break
1207
1208                self._advance(alnum=True)
1209
1210                # Nested comments are allowed by some dialects, e.g. databricks, duckdb, postgres
1211                if (
1212                    self.NESTED_COMMENTS
1213                    and not self._end
1214                    and self._chars(comment_end_size) == comment_start
1215                ):
1216                    self._advance(comment_start_size)
1217                    comment_count += 1
1218
1219            self._comments.append(self._text[comment_start_size : -comment_end_size + 1])
1220            self._advance(comment_end_size - 1)
1221        else:
1222            while not self._end and self.WHITE_SPACE.get(self._peek) is not TokenType.BREAK:
1223                self._advance(alnum=True)
1224            self._comments.append(self._text[comment_start_size:])
1225
1226        # Leading comment is attached to the succeeding token, whilst trailing comment to the preceding.
1227        # Multiple consecutive comments are preserved by appending them to the current comments list.
1228        if comment_start_line == self._prev_token_line:
1229            self.tokens[-1].comments.extend(self._comments)
1230            self._comments = []
1231            self._prev_token_line = self._line
1232
1233        return True
1234
1235    def _scan_number(self) -> None:
1236        if self._char == "0":
1237            peek = self._peek.upper()
1238            if peek == "B":
1239                return self._scan_bits() if self.BIT_STRINGS else self._add(TokenType.NUMBER)
1240            elif peek == "X":
1241                return self._scan_hex() if self.HEX_STRINGS else self._add(TokenType.NUMBER)
1242
1243        decimal = False
1244        scientific = 0
1245
1246        while True:
1247            if self._peek.isdigit():
1248                self._advance()
1249            elif self._peek == "." and not decimal:
1250                if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER:
1251                    return self._add(TokenType.NUMBER)
1252                decimal = True
1253                self._advance()
1254            elif self._peek in ("-", "+") and scientific == 1:
1255                scientific += 1
1256                self._advance()
1257            elif self._peek.upper() == "E" and not scientific:
1258                scientific += 1
1259                self._advance()
1260            elif self._peek.isidentifier():
1261                number_text = self._text
1262                literal = ""
1263
1264                while self._peek.strip() and self._peek not in self.SINGLE_TOKENS:
1265                    literal += self._peek
1266                    self._advance()
1267
1268                token_type = self.KEYWORDS.get(self.NUMERIC_LITERALS.get(literal.upper(), ""))
1269
1270                if token_type:
1271                    self._add(TokenType.NUMBER, number_text)
1272                    self._add(TokenType.DCOLON, "::")
1273                    return self._add(token_type, literal)
1274                elif self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT:
1275                    return self._add(TokenType.VAR)
1276
1277                self._advance(-len(literal))
1278                return self._add(TokenType.NUMBER, number_text)
1279            else:
1280                return self._add(TokenType.NUMBER)
1281
1282    def _scan_bits(self) -> None:
1283        self._advance()
1284        value = self._extract_value()
1285        try:
1286            # If `value` can't be converted to a binary, fallback to tokenizing it as an identifier
1287            int(value, 2)
1288            self._add(TokenType.BIT_STRING, value[2:])  # Drop the 0b
1289        except ValueError:
1290            self._add(TokenType.IDENTIFIER)
1291
1292    def _scan_hex(self) -> None:
1293        self._advance()
1294        value = self._extract_value()
1295        try:
1296            # If `value` can't be converted to a hex, fallback to tokenizing it as an identifier
1297            int(value, 16)
1298            self._add(TokenType.HEX_STRING, value[2:])  # Drop the 0x
1299        except ValueError:
1300            self._add(TokenType.IDENTIFIER)
1301
1302    def _extract_value(self) -> str:
1303        while True:
1304            char = self._peek.strip()
1305            if char and char not in self.SINGLE_TOKENS:
1306                self._advance(alnum=True)
1307            else:
1308                break
1309
1310        return self._text
1311
1312    def _scan_string(self, start: str) -> bool:
1313        base = None
1314        token_type = TokenType.STRING
1315
1316        if start in self._QUOTES:
1317            end = self._QUOTES[start]
1318        elif start in self._FORMAT_STRINGS:
1319            end, token_type = self._FORMAT_STRINGS[start]
1320
1321            if token_type == TokenType.HEX_STRING:
1322                base = 16
1323            elif token_type == TokenType.BIT_STRING:
1324                base = 2
1325            elif token_type == TokenType.HEREDOC_STRING:
1326                self._advance()
1327
1328                if self._char == end:
1329                    tag = ""
1330                else:
1331                    tag = self._extract_string(
1332                        end,
1333                        raw_string=True,
1334                        raise_unmatched=not self.HEREDOC_TAG_IS_IDENTIFIER,
1335                    )
1336
1337                if tag and self.HEREDOC_TAG_IS_IDENTIFIER and (self._end or not tag.isidentifier()):
1338                    if not self._end:
1339                        self._advance(-1)
1340
1341                    self._advance(-len(tag))
1342                    self._add(self.HEREDOC_STRING_ALTERNATIVE)
1343                    return True
1344
1345                end = f"{start}{tag}{end}"
1346        else:
1347            return False
1348
1349        self._advance(len(start))
1350        text = self._extract_string(end, raw_string=token_type == TokenType.RAW_STRING)
1351
1352        if base:
1353            try:
1354                int(text, base)
1355            except Exception:
1356                raise TokenError(
1357                    f"Numeric string contains invalid characters from {self._line}:{self._start}"
1358                )
1359
1360        self._add(token_type, text)
1361        return True
1362
1363    def _scan_identifier(self, identifier_end: str) -> None:
1364        self._advance()
1365        text = self._extract_string(identifier_end, escapes=self._IDENTIFIER_ESCAPES)
1366        self._add(TokenType.IDENTIFIER, text)
1367
1368    def _scan_var(self) -> None:
1369        while True:
1370            char = self._peek.strip()
1371            if char and (char in self.VAR_SINGLE_TOKENS or char not in self.SINGLE_TOKENS):
1372                self._advance(alnum=True)
1373            else:
1374                break
1375
1376        self._add(
1377            TokenType.VAR
1378            if self.tokens and self.tokens[-1].token_type == TokenType.PARAMETER
1379            else self.KEYWORDS.get(self._text.upper(), TokenType.VAR)
1380        )
1381
1382    def _extract_string(
1383        self,
1384        delimiter: str,
1385        escapes: t.Optional[t.Set[str]] = None,
1386        raw_string: bool = False,
1387        raise_unmatched: bool = True,
1388    ) -> str:
1389        text = ""
1390        delim_size = len(delimiter)
1391        escapes = self._STRING_ESCAPES if escapes is None else escapes
1392
1393        while True:
1394            if (
1395                not raw_string
1396                and self.dialect.UNESCAPED_SEQUENCES
1397                and self._peek
1398                and self._char in self.STRING_ESCAPES
1399            ):
1400                unescaped_sequence = self.dialect.UNESCAPED_SEQUENCES.get(self._char + self._peek)
1401                if unescaped_sequence:
1402                    self._advance(2)
1403                    text += unescaped_sequence
1404                    continue
1405            if (
1406                (self.STRING_ESCAPES_ALLOWED_IN_RAW_STRINGS or not raw_string)
1407                and self._char in escapes
1408                and (self._peek == delimiter or self._peek in escapes)
1409                and (self._char not in self._QUOTES or self._char == self._peek)
1410            ):
1411                if self._peek == delimiter:
1412                    text += self._peek
1413                else:
1414                    text += self._char + self._peek
1415
1416                if self._current + 1 < self.size:
1417                    self._advance(2)
1418                else:
1419                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._current}")
1420            else:
1421                if self._chars(delim_size) == delimiter:
1422                    if delim_size > 1:
1423                        self._advance(delim_size - 1)
1424                    break
1425
1426                if self._end:
1427                    if not raise_unmatched:
1428                        return text + self._char
1429
1430                    raise TokenError(f"Missing {delimiter} from {self._line}:{self._start}")
1431
1432                current = self._current - 1
1433                self._advance(alnum=True)
1434                text += self.sql[current : self._current - 1]
1435
1436        return text
1437
1438    def tokenize_rs(self, sql: str) -> t.List[Token]:
1439        if not self._RS_TOKENIZER:
1440            raise SqlglotError("Rust tokenizer is not available")
1441
1442        try:
1443            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1444            for token in tokens:
1445                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1446            return tokens
1447        except Exception as e:
1448            raise TokenError(str(e))
Tokenizer( dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None)
969    def __init__(self, dialect: DialectType = None) -> None:
970        from sqlglot.dialects import Dialect
971
972        self.dialect = Dialect.get_or_raise(dialect)
973
974        if USE_RS_TOKENIZER:
975            self._rs_dialect_settings = RsTokenizerDialectSettings(
976                unescaped_sequences=self.dialect.UNESCAPED_SEQUENCES,
977                identifiers_can_start_with_digit=self.dialect.IDENTIFIERS_CAN_START_WITH_DIGIT,
978            )
979
980        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'>, '~~*': <TokenType.ILIKE: 'ILIKE'>, '~*': <TokenType.IRLIKE: 'IRLIKE'>, '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.RENAME: 'RENAME'>, <TokenType.EXECUTE: 'EXECUTE'>, <TokenType.SHOW: 'SHOW'>, <TokenType.FETCH: 'FETCH'>, <TokenType.COMMAND: 'COMMAND'>}
COMMAND_PREFIX_TOKENS = {<TokenType.SEMICOLON: 'SEMICOLON'>, <TokenType.BEGIN: 'BEGIN'>}
NUMERIC_LITERALS: Dict[str, str] = {}
COMMENTS = ['--', ('/*', '*/')]
dialect
def reset(self) -> None:
982    def reset(self) -> None:
983        self.sql = ""
984        self.size = 0
985        self.tokens: t.List[Token] = []
986        self._start = 0
987        self._current = 0
988        self._line = 1
989        self._col = 0
990        self._comments: t.List[str] = []
991
992        self._char = ""
993        self._end = False
994        self._peek = ""
995        self._prev_token_line = -1
def tokenize(self, sql: str) -> List[Token]:
 997    def tokenize(self, sql: str) -> t.List[Token]:
 998        """Returns a list of tokens corresponding to the SQL string `sql`."""
 999        if USE_RS_TOKENIZER:
1000            return self.tokenize_rs(sql)
1001
1002        self.reset()
1003        self.sql = sql
1004        self.size = len(sql)
1005
1006        try:
1007            self._scan()
1008        except Exception as e:
1009            start = max(self._current - 50, 0)
1010            end = min(self._current + 50, self.size - 1)
1011            context = self.sql[start:end]
1012            raise TokenError(f"Error tokenizing '{context}'") from e
1013
1014        return self.tokens

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

def tokenize_rs(self, sql: str) -> List[Token]:
1438    def tokenize_rs(self, sql: str) -> t.List[Token]:
1439        if not self._RS_TOKENIZER:
1440            raise SqlglotError("Rust tokenizer is not available")
1441
1442        try:
1443            tokens = self._RS_TOKENIZER.tokenize(sql, self._rs_dialect_settings)
1444            for token in tokens:
1445                token.token_type = _ALL_TOKEN_TYPES[token.token_type_index]
1446            return tokens
1447        except Exception as e:
1448            raise TokenError(str(e))
size
sql
tokens