Lexer.php 29.1 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943
<?php

/**
 * Defines the lexer of the library.
 *
 * This is one of the most important components, along with the parser.
 *
 * Depends on context to extract lexemes.
 */

namespace PhpMyAdmin\SqlParser;

use PhpMyAdmin\SqlParser\Exceptions\LexerException;

if (!defined('USE_UTF_STRINGS')) {
    // NOTE: In previous versions of PHP (5.5 and older) the default
    // internal encoding is "ISO-8859-1".
    // All `mb_` functions must specify the correct encoding, which is
    // 'UTF-8' in order to work properly.

    /*
     * Forces usage of `UtfString` if the string is multibyte.
     * `UtfString` may be slower, but it gives better results.
     *
     * @var bool
     */
    define('USE_UTF_STRINGS', true);
}

/**
 * Performs lexical analysis over a SQL statement and splits it in multiple
 * tokens.
 *
 * The output of the lexer is affected by the context of the SQL statement.
 *
 * @category Lexer
 *
 * @license  https://www.gnu.org/licenses/gpl-2.0.txt GPL-2.0+
 *
 * @see      Context
 */
class Lexer extends Core
{
    /**
     * A list of methods that are used in lexing the SQL query.
     *
     * @var array
     */
    public static $PARSER_METHODS = array(
        // It is best to put the parsers in order of their complexity
        // (ascending) and their occurrence rate (descending).
        //
        // Conflicts:
        //
        // 1. `parseDelimiter`, `parseUnknown`, `parseKeyword`, `parseNumber`
        // They fight over delimiter. The delimiter may be a keyword, a
        // number or almost any character which makes the delimiter one of
        // the first tokens that must be parsed.
        //
        // 1. `parseNumber` and `parseOperator`
        // They fight over `+` and `-`.
        //
        // 2. `parseComment` and `parseOperator`
        // They fight over `/` (as in ```/*comment*/``` or ```a / b```)
        //
        // 3. `parseBool` and `parseKeyword`
        // They fight over `TRUE` and `FALSE`.
        //
        // 4. `parseKeyword` and `parseUnknown`
        // They fight over words. `parseUnknown` does not know about
        // keywords.

        'parseDelimiter', 'parseWhitespace', 'parseNumber',
        'parseComment', 'parseOperator', 'parseBool', 'parseString',
        'parseSymbol', 'parseKeyword', 'parseLabel', 'parseUnknown',
    );

    /**
     * The string to be parsed.
     *
     * @var string|UtfString
     */
    public $str = '';

    /**
     * The length of `$str`.
     *
     * By storing its length, a lot of time is saved, because parsing methods
     * would call `strlen` everytime.
     *
     * @var int
     */
    public $len = 0;

    /**
     * The index of the last parsed character.
     *
     * @var int
     */
    public $last = 0;

    /**
     * Tokens extracted from given strings.
     *
     * @var TokensList
     */
    public $list;

    /**
     * The default delimiter. This is used, by default, in all new instances.
     *
     * @var string
     */
    public static $DEFAULT_DELIMITER = ';';

    /**
     * Statements delimiter.
     * This may change during lexing.
     *
     * @var string
     */
    public $delimiter;

    /**
     * The length of the delimiter.
     *
     * Because `parseDelimiter` can be called a lot, it would perform a lot of
     * calls to `strlen`, which might affect performance when the delimiter is
     * big.
     *
     * @var int
     */
    public $delimiterLen;

    /**
     * Gets the tokens list parsed by a new instance of a lexer.
     *
     * @param string|UtfString $str       the query to be lexed
     * @param bool             $strict    whether strict mode should be
     *                                    enabled or not
     * @param string           $delimiter the delimiter to be used
     *
     * @return TokensList
     */
    public static function getTokens($str, $strict = false, $delimiter = null)
    {
        $lexer = new self($str, $strict, $delimiter);

        return $lexer->list;
    }

    /**
     * Constructor.
     *
     * @param string|UtfString $str       the query to be lexed
     * @param bool             $strict    whether strict mode should be
     *                                    enabled or not
     * @param string           $delimiter the delimiter to be used
     */
    public function __construct($str, $strict = false, $delimiter = null)
    {
        // `strlen` is used instead of `mb_strlen` because the lexer needs to
        // parse each byte of the input.
        $len = $str instanceof UtfString ? $str->length() : strlen($str);

        // For multi-byte strings, a new instance of `UtfString` is
        // initialized (only if `UtfString` usage is forced.
        if (!$str instanceof UtfString && USE_UTF_STRINGS && $len !== mb_strlen($str, 'UTF-8')) {
            $str = new UtfString($str);
        }

        $this->str = $str;
        $this->len = $str instanceof UtfString ? $str->length() : $len;

        $this->strict = $strict;

        // Setting the delimiter.
        $this->setDelimiter(
            !empty($delimiter) ? $delimiter : static::$DEFAULT_DELIMITER
        );

        $this->lex();
    }

    /**
     * Sets the delimiter.
     *
     * @param string $delimiter the new delimiter
     */
    public function setDelimiter($delimiter)
    {
        $this->delimiter = $delimiter;
        $this->delimiterLen = strlen($delimiter);
    }

    /**
     * Parses the string and extracts lexemes.
     */
    public function lex()
    {
        // TODO: Sometimes, static::parse* functions make unnecessary calls to
        // is* functions. For a better performance, some rules can be deduced
        // from context.
        // For example, in `parseBool` there is no need to compare the token
        // every time with `true` and `false`. The first step would be to
        // compare with 'true' only and just after that add another letter from
        // context and compare again with `false`.
        // Another example is `parseComment`.

        $list = new TokensList();

        /**
         * Last processed token.
         *
         * @var Token
         */
        $lastToken = null;

        for ($this->last = 0, $lastIdx = 0; $this->last < $this->len; $lastIdx = ++$this->last) {
            /**
             * The new token.
             *
             * @var Token
             */
            $token = null;

            foreach (static::$PARSER_METHODS as $method) {
                if ($token = $this->$method()) {
                    break;
                }
            }

            if ($token === null) {
                // @assert($this->last === $lastIdx);
                $token = new Token($this->str[$this->last]);
                $this->error(
                    'Unexpected character.',
                    $this->str[$this->last],
                    $this->last
                );
            } elseif ($lastToken !== null
                && $token->type === Token::TYPE_SYMBOL
                && $token->flags & Token::FLAG_SYMBOL_VARIABLE
                && (
                    $lastToken->type === Token::TYPE_STRING
                    || (
                        $lastToken->type === Token::TYPE_SYMBOL
                        && $lastToken->flags & Token::FLAG_SYMBOL_BACKTICK
                    )
                )
            ) {
                // Handles ```... FROM 'user'@'%' ...```.
                $lastToken->token .= $token->token;
                $lastToken->type = Token::TYPE_SYMBOL;
                $lastToken->flags = Token::FLAG_SYMBOL_USER;
                $lastToken->value .= '@' . $token->value;
                continue;
            } elseif ($lastToken !== null
                && $token->type === Token::TYPE_KEYWORD
                && $lastToken->type === Token::TYPE_OPERATOR
                && $lastToken->value === '.'
            ) {
                // Handles ```... tbl.FROM ...```. In this case, FROM is not
                // a reserved word.
                $token->type = Token::TYPE_NONE;
                $token->flags = 0;
                $token->value = $token->token;
            }

            $token->position = $lastIdx;

            $list->tokens[$list->count++] = $token;

            // Handling delimiters.
            if ($token->type === Token::TYPE_NONE && $token->value === 'DELIMITER') {
                if ($this->last + 1 >= $this->len) {
                    $this->error(
                        'Expected whitespace(s) before delimiter.',
                        '',
                        $this->last + 1
                    );
                    continue;
                }

                // Skipping last R (from `delimiteR`) and whitespaces between
                // the keyword `DELIMITER` and the actual delimiter.
                $pos = ++$this->last;
                if (($token = $this->parseWhitespace()) !== null) {
                    $token->position = $pos;
                    $list->tokens[$list->count++] = $token;
                }

                // Preparing the token that holds the new delimiter.
                if ($this->last + 1 >= $this->len) {
                    $this->error(
                        'Expected delimiter.',
                        '',
                        $this->last + 1
                    );
                    continue;
                }
                $pos = $this->last + 1;

                // Parsing the delimiter.
                $this->delimiter = null;
                $delimiterLen = 0;
                while (++$this->last < $this->len && !Context::isWhitespace($this->str[$this->last]) && $delimiterLen < 15) {
                    $this->delimiter .= $this->str[$this->last];
                    ++$delimiterLen;
                }

                if (empty($this->delimiter)) {
                    $this->error(
                        'Expected delimiter.',
                        '',
                        $this->last
                    );
                    $this->delimiter = ';';
                }

                --$this->last;

                // Saving the delimiter and its token.
                $this->delimiterLen = strlen($this->delimiter);
                $token = new Token($this->delimiter, Token::TYPE_DELIMITER);
                $token->position = $pos;
                $list->tokens[$list->count++] = $token;
            }

            $lastToken = $token;
        }

        // Adding a final delimiter to mark the ending.
        $list->tokens[$list->count++] = new Token(null, Token::TYPE_DELIMITER);

        // Saving the tokens list.
        $this->list = $list;
    }

    /**
     * Creates a new error log.
     *
     * @param string $msg  the error message
     * @param string $str  the character that produced the error
     * @param int    $pos  the position of the character
     * @param int    $code the code of the error
     *
     * @throws LexerException throws the exception, if strict mode is enabled
     */
    public function error($msg, $str = '', $pos = 0, $code = 0)
    {
        $error = new LexerException(
            Translator::gettext($msg),
            $str, $pos, $code
        );
        parent::error($error);
    }

    /**
     * Parses a keyword.
     *
     * @return null|Token
     */
    public function parseKeyword()
    {
        $token = '';

        /**
         * Value to be returned.
         *
         * @var Token
         */
        $ret = null;

        /**
         * The value of `$this->last` where `$token` ends in `$this->str`.
         *
         * @var int
         */
        $iEnd = $this->last;

        /**
         * Whether last parsed character is a whitespace.
         *
         * @var bool
         */
        $lastSpace = false;

        for ($j = 1; $j < Context::KEYWORD_MAX_LENGTH && $this->last < $this->len; ++$j, ++$this->last) {
            // Composed keywords shouldn't have more than one whitespace between
            // keywords.
            if (Context::isWhitespace($this->str[$this->last])) {
                if ($lastSpace) {
                    --$j; // The size of the keyword didn't increase.
                    continue;
                }
                $lastSpace = true;
            } else {
                $lastSpace = false;
            }

            $token .= $this->str[$this->last];
            if (($this->last + 1 === $this->len || Context::isSeparator($this->str[$this->last + 1]))
                && $flags = Context::isKeyword($token)
            ) {
                $ret = new Token($token, Token::TYPE_KEYWORD, $flags);
                $iEnd = $this->last;

                // We don't break so we find longest keyword.
                // For example, `OR` and `ORDER` have a common prefix `OR`.
                // If we stopped at `OR`, the parsing would be invalid.
            }
        }

        $this->last = $iEnd;

        return $ret;
    }

    /**
     * Parses a label.
     *
     * @return null|Token
     */
    public function parseLabel()
    {
        $token = '';

        /**
         * Value to be returned.
         *
         * @var Token
         */
        $ret = null;

        /**
         * The value of `$this->last` where `$token` ends in `$this->str`.
         *
         * @var int
         */
        $iEnd = $this->last;
        for ($j = 1; $j < Context::LABEL_MAX_LENGTH && $this->last < $this->len; ++$j, ++$this->last) {
            if ($this->str[$this->last] === ':' && $j > 1) {
                // End of label
                $token .= $this->str[$this->last];
                $ret = new Token($token, Token::TYPE_LABEL);
                $iEnd = $this->last;
                break;
            } elseif (Context::isWhitespace($this->str[$this->last]) && $j > 1) {
                // Whitespace between label and :
                // The size of the keyword didn't increase.
                --$j;
            } elseif (Context::isSeparator($this->str[$this->last])) {
                // Any other separator
                break;
            }
            $token .= $this->str[$this->last];
        }

        $this->last = $iEnd;

        return $ret;
    }

    /**
     * Parses an operator.
     *
     * @return null|Token
     */
    public function parseOperator()
    {
        $token = '';

        /**
         * Value to be returned.
         *
         * @var Token
         */
        $ret = null;

        /**
         * The value of `$this->last` where `$token` ends in `$this->str`.
         *
         * @var int
         */
        $iEnd = $this->last;

        for ($j = 1; $j < Context::OPERATOR_MAX_LENGTH && $this->last < $this->len; ++$j, ++$this->last) {
            $token .= $this->str[$this->last];
            if ($flags = Context::isOperator($token)) {
                $ret = new Token($token, Token::TYPE_OPERATOR, $flags);
                $iEnd = $this->last;
            }
        }

        $this->last = $iEnd;

        return $ret;
    }

    /**
     * Parses a whitespace.
     *
     * @return null|Token
     */
    public function parseWhitespace()
    {
        $token = $this->str[$this->last];

        if (!Context::isWhitespace($token)) {
            return null;
        }

        while (++$this->last < $this->len && Context::isWhitespace($this->str[$this->last])) {
            $token .= $this->str[$this->last];
        }

        --$this->last;

        return new Token($token, Token::TYPE_WHITESPACE);
    }

    /**
     * Parses a comment.
     *
     * @return null|Token
     */
    public function parseComment()
    {
        $iBak = $this->last;
        $token = $this->str[$this->last];

        // Bash style comments. (#comment\n)
        if (Context::isComment($token)) {
            while (
                ++$this->last < $this->len
                && $this->str[$this->last] !== "\n"
            ) {
                $token .= $this->str[$this->last];
            }
            // Include trailing \n as whitespace token
            if ($this->last < $this->len) {
                --$this->last;
            }

            return new Token($token, Token::TYPE_COMMENT, Token::FLAG_COMMENT_BASH);
        }

        // C style comments. (/*comment*\/)
        if (++$this->last < $this->len) {
            $token .= $this->str[$this->last];
            if (Context::isComment($token)) {
                $flags = Token::FLAG_COMMENT_C;

                // This comment already ended. It may be a part of a
                // previous MySQL specific command.
                if ($token === '*/') {
                    return new Token($token, Token::TYPE_COMMENT, $flags);
                }

                // Checking if this is a MySQL-specific command.
                if ($this->last + 1 < $this->len
                    && $this->str[$this->last + 1] === '!'
                ) {
                    $flags |= Token::FLAG_COMMENT_MYSQL_CMD;
                    $token .= $this->str[++$this->last];

                    while (
                        ++$this->last < $this->len
                        && $this->str[$this->last] >= '0'
                        && $this->str[$this->last] <= '9'
                    ) {
                        $token .= $this->str[$this->last];
                    }
                    --$this->last;

                    // We split this comment and parse only its beginning
                    // here.
                    return new Token($token, Token::TYPE_COMMENT, $flags);
                }

                // Parsing the comment.
                while (
                    ++$this->last < $this->len
                    && (
                        $this->str[$this->last - 1] !== '*'
                        || $this->str[$this->last] !== '/'
                    )
                ) {
                    $token .= $this->str[$this->last];
                }

                // Adding the ending.
                if ($this->last < $this->len) {
                    $token .= $this->str[$this->last];
                }

                return new Token($token, Token::TYPE_COMMENT, $flags);
            }
        }

        // SQL style comments. (-- comment\n)
        if (++$this->last < $this->len) {
            $token .= $this->str[$this->last];
            $end = false;
        } else {
            --$this->last;
            $end = true;
        }
        if (Context::isComment($token, $end)) {
            // Checking if this comment did not end already (```--\n```).
            if ($this->str[$this->last] !== "\n") {
                while (
                    ++$this->last < $this->len
                    && $this->str[$this->last] !== "\n"
                ) {
                    $token .= $this->str[$this->last];
                }
            }
            // Include trailing \n as whitespace token
            if ($this->last < $this->len) {
                --$this->last;
            }

            return new Token($token, Token::TYPE_COMMENT, Token::FLAG_COMMENT_SQL);
        }

        $this->last = $iBak;

        return null;
    }

    /**
     * Parses a boolean.
     *
     * @return null|Token
     */
    public function parseBool()
    {
        if ($this->last + 3 >= $this->len) {
            // At least `min(strlen('TRUE'), strlen('FALSE'))` characters are
            // required.
            return null;
        }

        $iBak = $this->last;
        $token = $this->str[$this->last] . $this->str[++$this->last]
        . $this->str[++$this->last] . $this->str[++$this->last]; // _TRUE_ or _FALS_e

        if (Context::isBool($token)) {
            return new Token($token, Token::TYPE_BOOL);
        } elseif (++$this->last < $this->len) {
            $token .= $this->str[$this->last]; // fals_E_
            if (Context::isBool($token)) {
                return new Token($token, Token::TYPE_BOOL, 1);
            }
        }

        $this->last = $iBak;

        return null;
    }

    /**
     * Parses a number.
     *
     * @return null|Token
     */
    public function parseNumber()
    {
        // A rudimentary state machine is being used to parse numbers due to
        // the various forms of their notation.
        //
        // Below are the states of the machines and the conditions to change
        // the state.
        //
        //      1 --------------------[ + or - ]-------------------> 1
        //      1 -------------------[ 0x or 0X ]------------------> 2
        //      1 --------------------[ 0 to 9 ]-------------------> 3
        //      1 -----------------------[ . ]---------------------> 4
        //      1 -----------------------[ b ]---------------------> 7
        //
        //      2 --------------------[ 0 to F ]-------------------> 2
        //
        //      3 --------------------[ 0 to 9 ]-------------------> 3
        //      3 -----------------------[ . ]---------------------> 4
        //      3 --------------------[ e or E ]-------------------> 5
        //
        //      4 --------------------[ 0 to 9 ]-------------------> 4
        //      4 --------------------[ e or E ]-------------------> 5
        //
        //      5 ---------------[ + or - or 0 to 9 ]--------------> 6
        //
        //      7 -----------------------[ ' ]---------------------> 8
        //
        //      8 --------------------[ 0 or 1 ]-------------------> 8
        //      8 -----------------------[ ' ]---------------------> 9
        //
        // State 1 may be reached by negative numbers.
        // State 2 is reached only by hex numbers.
        // State 4 is reached only by float numbers.
        // State 5 is reached only by numbers in approximate form.
        // State 7 is reached only by numbers in bit representation.
        //
        // Valid final states are: 2, 3, 4 and 6. Any parsing that finished in a
        // state other than these is invalid.
        $iBak = $this->last;
        $token = '';
        $flags = 0;
        $state = 1;
        for (; $this->last < $this->len; ++$this->last) {
            if ($state === 1) {
                if ($this->str[$this->last] === '-') {
                    $flags |= Token::FLAG_NUMBER_NEGATIVE;
                } elseif ($this->last + 1 < $this->len
                    && $this->str[$this->last] === '0'
                    && (
                        $this->str[$this->last + 1] === 'x'
                        || $this->str[$this->last + 1] === 'X'
                    )
                ) {
                    $token .= $this->str[$this->last++];
                    $state = 2;
                } elseif ($this->str[$this->last] >= '0' && $this->str[$this->last] <= '9') {
                    $state = 3;
                } elseif ($this->str[$this->last] === '.') {
                    $state = 4;
                } elseif ($this->str[$this->last] === 'b') {
                    $state = 7;
                } elseif ($this->str[$this->last] !== '+') {
                    // `+` is a valid character in a number.
                    break;
                }
            } elseif ($state === 2) {
                $flags |= Token::FLAG_NUMBER_HEX;
                if (
                    !(
                        ($this->str[$this->last] >= '0' && $this->str[$this->last] <= '9')
                        || ($this->str[$this->last] >= 'A' && $this->str[$this->last] <= 'F')
                        || ($this->str[$this->last] >= 'a' && $this->str[$this->last] <= 'f')
                    )
                ) {
                    break;
                }
            } elseif ($state === 3) {
                if ($this->str[$this->last] === '.') {
                    $state = 4;
                } elseif ($this->str[$this->last] === 'e' || $this->str[$this->last] === 'E') {
                    $state = 5;
                } elseif ($this->str[$this->last] < '0' || $this->str[$this->last] > '9') {
                    // Just digits and `.`, `e` and `E` are valid characters.
                    break;
                }
            } elseif ($state === 4) {
                $flags |= Token::FLAG_NUMBER_FLOAT;
                if ($this->str[$this->last] === 'e' || $this->str[$this->last] === 'E') {
                    $state = 5;
                } elseif ($this->str[$this->last] < '0' || $this->str[$this->last] > '9') {
                    // Just digits, `e` and `E` are valid characters.
                    break;
                }
            } elseif ($state === 5) {
                $flags |= Token::FLAG_NUMBER_APPROXIMATE;
                if ($this->str[$this->last] === '+' || $this->str[$this->last] === '-'
                    || ($this->str[$this->last] >= '0' && $this->str[$this->last] <= '9')
                ) {
                    $state = 6;
                } else {
                    break;
                }
            } elseif ($state === 6) {
                if ($this->str[$this->last] < '0' || $this->str[$this->last] > '9') {
                    // Just digits are valid characters.
                    break;
                }
            } elseif ($state === 7) {
                $flags |= Token::FLAG_NUMBER_BINARY;
                if ($this->str[$this->last] === '\'') {
                    $state = 8;
                } else {
                    break;
                }
            } elseif ($state === 8) {
                if ($this->str[$this->last] === '\'') {
                    $state = 9;
                } elseif ($this->str[$this->last] !== '0'
                    && $this->str[$this->last] !== '1'
                ) {
                    break;
                }
            } elseif ($state === 9) {
                break;
            }
            $token .= $this->str[$this->last];
        }
        if ($state === 2 || $state === 3
            || ($token !== '.' && $state === 4)
            || $state === 6 || $state === 9
        ) {
            --$this->last;

            return new Token($token, Token::TYPE_NUMBER, $flags);
        }
        $this->last = $iBak;

        return null;
    }

    /**
     * Parses a string.
     *
     * @param string $quote additional starting symbol
     *
     * @return null|Token
     */
    public function parseString($quote = '')
    {
        $token = $this->str[$this->last];
        if (!($flags = Context::isString($token)) && $token !== $quote) {
            return null;
        }
        $quote = $token;

        while (++$this->last < $this->len) {
            if ($this->last + 1 < $this->len
                && (
                    ($this->str[$this->last] === $quote && $this->str[$this->last + 1] === $quote)
                    || ($this->str[$this->last] === '\\' && $quote !== '`')
                )
            ) {
                $token .= $this->str[$this->last] . $this->str[++$this->last];
            } else {
                if ($this->str[$this->last] === $quote) {
                    break;
                }
                $token .= $this->str[$this->last];
            }
        }

        if ($this->last >= $this->len || $this->str[$this->last] !== $quote) {
            $this->error(
                sprintf(
                    Translator::gettext('Ending quote %1$s was expected.'),
                    $quote
                ),
                '',
                $this->last
            );
        } else {
            $token .= $this->str[$this->last];
        }

        return new Token($token, Token::TYPE_STRING, $flags);
    }

    /**
     * Parses a symbol.
     *
     * @return null|Token
     */
    public function parseSymbol()
    {
        $token = $this->str[$this->last];
        if (!($flags = Context::isSymbol($token))) {
            return null;
        }

        if ($flags & Token::FLAG_SYMBOL_VARIABLE) {
            if ($this->last + 1 < $this->len && $this->str[++$this->last] === '@') {
                // This is a system variable (e.g. `@@hostname`).
                $token .= $this->str[$this->last++];
                $flags |= Token::FLAG_SYMBOL_SYSTEM;
            }
        } elseif ($flags & Token::FLAG_SYMBOL_PARAMETER) {
            if ($this->last + 1 < $this->len) {
                ++$this->last;
            }
        } else {
            $token = '';
        }

        $str = null;

        if ($this->last < $this->len) {
            if (($str = $this->parseString('`')) === null) {
                if (($str = $this->parseUnknown()) === null) {
                    $this->error(
                        'Variable name was expected.',
                        $this->str[$this->last],
                        $this->last
                    );
                }
            }
        }

        if ($str !== null) {
            $token .= $str->token;
        }

        return new Token($token, Token::TYPE_SYMBOL, $flags);
    }

    /**
     * Parses unknown parts of the query.
     *
     * @return null|Token
     */
    public function parseUnknown()
    {
        $token = $this->str[$this->last];
        if (Context::isSeparator($token)) {
            return null;
        }

        while (++$this->last < $this->len && !Context::isSeparator($this->str[$this->last])) {
            $token .= $this->str[$this->last];
        }
        --$this->last;

        return new Token($token);
    }

    /**
     * Parses the delimiter of the query.
     *
     * @return null|Token
     */
    public function parseDelimiter()
    {
        $idx = 0;

        while ($idx < $this->delimiterLen && $this->last + $idx < $this->len) {
            if ($this->delimiter[$idx] !== $this->str[$this->last + $idx]) {
                return null;
            }
            ++$idx;
        }

        $this->last += $this->delimiterLen - 1;

        return new Token($this->delimiter, Token::TYPE_DELIMITER);
    }
}