Lexer::readToken() - Code Metrics - Inspection of "Lexer optimizations" - digiaonline/graphql-php - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Pull Request — master (#196)

by Christoffer

created 2018-04-11 20:52 UTC

Lexer::readToken() C

↳ Parent: Lexer

Complexity

Conditions	24
Paths	9

Size

Total Lines	53
Code Lines	23

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
dl	0
loc	53
rs	6.1349
c	0
b	0
f	0
cc	24
eloc	23
nc	9
nop	1

How to fix Long Method Complexity

<?php

namespace Digia\GraphQL\Language;

use Digia\GraphQL\Error\SyntaxErrorException;

class Lexer implements LexerInterface
{
    protected const ENCODING = 'UTF-8';

    /**
     * A map between punctuation character code and the corresponding token kind.
     *
     * @var array
     */
    protected static $codeTokenKindMap = [
        33  => TokenKindEnum::BANG,
        36  => TokenKindEnum::DOLLAR,
        38  => TokenKindEnum::AMP,
        40  => TokenKindEnum::PAREN_L,
        41  => TokenKindEnum::PAREN_R,
        58  => TokenKindEnum::COLON,
        61  => TokenKindEnum::EQUALS,
        64  => TokenKindEnum::AT,
        91  => TokenKindEnum::BRACKET_L,
        93  => TokenKindEnum::BRACKET_R,
        123 => TokenKindEnum::BRACE_L,
        124 => TokenKindEnum::PIPE,
        125 => TokenKindEnum::BRACE_R,
    ];

    /**
     * The source file for this lexer.
     *
     * @var Source
     */
    protected $source;

    /**
     * The contents of the source file.
     *
     * @var string
     */
    protected $body;

    /**
     * The total number of characters in the source file.
     *
     * @var int
     */
    protected $bodyLength;

    /**
     * The options for this lexer.
     *
     * @var array
     */
    protected $options = [];

    /**
     * The previously focused non-ignored token.
     *
     * @var Token
     */
    protected $lastToken;

    /**
     * The currently focused non-ignored token.
     *
     * @var Token
     */
    protected $token;

    /**
     * The current position.
     *
     * @var int
     */
    protected $position;

    /**
     * The (1-indexed) line containing the current token.
     *
     * @var int
     */
    protected $line;

    /**
     * The character offset at which the current line begins.
     *
     * @var int
     */
    protected $lineStart;

    /**
     * A key-value map over characters and their corresponding character codes.
     *
     * @var array
     */
    protected static $charCodeCache = [];

    /**
     * Lexer constructor.
     * @param Source|null $source
     * @param array       $options
     */
    public function __construct(Source $source, array $options)
    {
        $startOfFileToken = $this->createStartOfFileToken();

        $this->lastToken  = $startOfFileToken;
        $this->token      = $startOfFileToken;
        $this->line       = 1;
        $this->lineStart  = 0;
        $this->body       = $source->getBody();
        $this->bodyLength = \strlen($this->body);
        $this->source     = $source;
        $this->options    = $options;
    }

    /**
     * @inheritdoc
     * @throws SyntaxErrorException
     */
    public function advance(): Token
    {
        $this->lastToken = $this->token;
        return $this->token = $this->lookahead();
    }

    /**
     * @inheritdoc
     * @throws SyntaxErrorException
     */
    public function lookahead(): Token
    {
        $token = $this->token;

        if (TokenKindEnum::EOF !== $token->getKind()) {
            do {
                $next = $this->readToken($token);
                $token->setNext($next);
                $token = $next;
            } while (TokenKindEnum::COMMENT === $token->getKind());
        }

        return $token;
    }

    /**
     * @inheritdoc
     */
    public function getOption(string $name, $default = null)
    {
        return $this->options[$name] ?? $default;
    }

    /**
     * @inheritdoc
     */
    public function getSource(): Source
    {
        return $this->source;
    }

    /**
     * @inheritdoc
     */
    public function getToken(): Token
    {
        return $this->token;
    }

    /**
     * @inheritdoc
     */
    public function getLastToken(): Token
    {
        return $this->lastToken;
    }

    /**
     * @inheritdoc
     */
    public function createSyntaxErrorException(?string $description = null): SyntaxErrorException
    {
        return new SyntaxErrorException(
            $this->source,
            $this->position,
            $description ?? $this->unexpectedCharacterMessage($this->readCharCode($this->position))
        );
    }

    /**
     * Reads the token after the given token.
     *
     * @param Token $prev
     * @return Token
     * @throws SyntaxErrorException
     */
    protected function readToken(Token $prev): Token
    {
        $this->position = $prev->getEnd();

        $this->skipWhitespace();

        $line   = $this->line;
        $column = (1 + $this->position) - $this->lineStart;

        if ($this->position >= $this->bodyLength) {
            return $this->createEndOfFileToken($line, $column, $prev);
        }

        $code = $this->readCharCode($this->position);

        // Punctuation: [!$&:=@|()\[\]{}]{1}
        if (33 === $code || 36 === $code || 38 === $code || 58 === $code || 61 === $code || 64 === $code || 124 === $code ||
            40 === $code || 41 === $code || 91 === $code || 93 === $code || 123 === $code || 125 === $code) {
            return $this->lexPunctuation($code, $line, $column, $prev);
        }

        // Comment: #[\u0009\u0020-\uFFFF]*
        if (35 === $code) {
            return $this->lexComment($line, $column, $prev);
        }

        // Int:   -?(0|[1-9][0-9]*)
        // Float: -?(0|[1-9][0-9]*)(\.[0-9]+)?((E|e)(+|-)?[0-9]+)?
        if (45 === $code || isNumber($code)) {
            return $this->lexNumber($code, $line, $column, $prev);
        }

        // Name: [_A-Za-z][_0-9A-Za-z]*
        if (isAlphaNumeric($code)) {
            return $this->lexName($line, $column, $prev);
        }

        // Spread: ...
        if ($this->bodyLength >= 3 && $this->isSpread($code)) {
            return $this->lexSpread($line, $column, $prev);
        }

        // String: "([^"\\\u000A\u000D]|(\\(u[0-9a-fA-F]{4}|["\\/bfnrt])))*"
        if ($this->isString($code)) {
            return $this->lexString($line, $column, $prev);
        }

        // Block String: """("?"?(\\"""|\\(?!=""")|[^"\\]))*"""
        if ($this->bodyLength >= 3 && $this->isTripleQuote($code)) {
            return $this->lexBlockString($line, $column, $prev);
        }

        throw $this->createSyntaxErrorException();
    }

    /**
     * @return Token
     */
    protected function createStartOfFileToken(): Token
    {
        return new Token(TokenKindEnum::SOF);
    }

    /**
     * Creates an End Of File (EOF) token.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     */
    protected function createEndOfFileToken(int $line, int $column, Token $prev): Token
    {
        return new Token(TokenKindEnum::EOF, $this->bodyLength, $this->bodyLength, $line, $column, $prev);
    }

    /**
     * Reads a punctuation token from the source file.
     *
     * @param int   $code
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     * @throws SyntaxErrorException
     */
    protected function lexPunctuation(int $code, int $line, int $column, Token $prev): ?Token
    {
        if (!isset(self::$codeTokenKindMap[$code])) {
            throw $this->createSyntaxErrorException();
        }

        return new Token(self::$codeTokenKindMap[$code], $this->position, $this->position + 1, $line, $column, $prev);
    }

    /**
     * Reads a name token from the source file.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     */
    protected function lexName(int $line, int $column, Token $prev): Token
    {
        $start = $this->position;

        ++$this->position;

        while ($this->position !== $this->bodyLength &&
            ($code = $this->readCharCode($this->position)) !== null &&
            isAlphaNumeric($code)) {
            ++$this->position;
        }

        $value = sliceString($this->body, $start, $this->position);

        return new Token(TokenKindEnum::NAME, $start, $this->position, $line, $column, $prev, $value);
    }

    /**
     * Reads a number (int or float) token from the source file.
     *
     * @param int   $code
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     * @throws SyntaxErrorException
     */
    protected function lexNumber(int $code, int $line, int $column, Token $prev): Token
    {
        $start   = $this->position;
        $isFloat = false;

        if (45 === $code) {
            // -
            $code = $this->readCharCode(++$this->position);
        }

        if (48 === $code) {
            // 0
            $code = $this->readCharCode(++$this->position);

            if (isNumber($code)) {
                throw $this->createSyntaxErrorException(
                    \sprintf('Invalid number, unexpected digit after 0: %s.', printCharCode($code))
                );
            }
        } else {
            $this->skipDigits($code);
            $code = $this->readCharCode($this->position);
        }

        if (46 === $code) {
            // .
            $isFloat = true;

            $code = $this->readCharCode(++$this->position);
            $this->skipDigits($code);
            $code = $this->readCharCode($this->position);
        }

        if (69 === $code || 101 === $code) {
            // e or E
            $isFloat = true;

            $code = $this->readCharCode(++$this->position);

            if (43 === $code || 45 === $code) {
                // + or -
                $code = $this->readCharCode(++$this->position);
            }

            $this->skipDigits($code);
        }

        return new Token(
            $isFloat ? TokenKindEnum::FLOAT : TokenKindEnum::INT,
            $start,
            $this->position,
            $line,
            $column,
            $prev,
            sliceString($this->body, $start, $this->position)
        );
    }

    /**
     * Skips digits at the current position.
     *
     * @param int $code
     * @throws SyntaxErrorException
     */
    protected function skipDigits(int $code): void
    {
        if (isNumber($code)) {
            do {
                $code = $this->readCharCode(++$this->position);
            } while (isNumber($code));

            return;
        }

        throw $this->createSyntaxErrorException(
            \sprintf('Invalid number, expected digit but got: %s.', printCharCode($code))
        );
    }

    /**
     * Reads a comment token from the source file.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     */
    protected function lexComment(int $line, int $column, Token $prev): Token
    {
        $start = $this->position;

        do {
            $code = $this->readCharCode(++$this->position);
        } while ($code !== null && ($code > 0x001f || 0x0009 === $code)); // SourceCharacter but not LineTerminator

        return new Token(
            TokenKindEnum::COMMENT,
            $start,
            $this->position,
            $line,
            $column,
            $prev,
            sliceString($this->body, $start + 1, $this->position)
        );
    }

    /**
     * Reads a spread token from the source.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     */
    protected function lexSpread(int $line, int $column, Token $prev): Token
    {
        return new Token(TokenKindEnum::SPREAD, $this->position, $this->position + 3, $line, $column, $prev);
    }

    /**
     * Reads a string token from the source.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     * @throws SyntaxErrorException
     */
    protected function lexString(int $line, int $column, Token $prev): Token
    {
        $start      = $this->position;
        $chunkStart = ++$this->position; // skip the quote
        $value      = '';

        while ($this->position < $this->bodyLength &&
            ($code = $this->readCharCode($this->position)) !== null && !isLineTerminator($code)) {
            // Closing Quote (")
            if (34 === $code) {
                $value .= sliceString($this->body, $chunkStart, $this->position);
                return new Token(TokenKindEnum::STRING, $start, $this->position + 1, $line, $column, $prev, $value);
            }

            if (isSourceCharacter($code)) {
                throw $this->createSyntaxErrorException(
                    \sprintf('Invalid character within String: %s.', printCharCode($code))
                );
            }

            ++$this->position;

            if (92 === $code) {
                // \
                $value .= sliceString($this->body, $chunkStart, $this->position - 1);

                $code = $this->readCharCode($this->position);

                switch ($code) {
                    case 34: // "
                        $value .= '"';
                        break;
                    case 47: // /
                        $value .= '/';
                        break;
                    case 92: // \
                        $value .= '\\';
                        break;
                    case 98: // b
                        $value .= '\b';
                        break;
                    case 102: // f
                        $value .= '\f';
                        break;
                    case 110: // n
                        $value .= '\n';
                        break;
                    case 114: // r
                        $value .= '\r';
                        break;
                    case 116: // t
                        $value .= '\t';
                        break;
                    case 117: // u
                        $unicodeString = sliceString($this->body, $this->position + 1, $this->position + 5);

                        if (!\preg_match('/[0-9A-Fa-f]{4}/', $unicodeString)) {
                            throw $this->createSyntaxErrorException(
                                \sprintf('Invalid character escape sequence: \\u%s.', $unicodeString)
                            );
                        }

                        $value .= '\\u' . $unicodeString;

                        $this->position += 4;

                        break;
                    default:
                        throw $this->createSyntaxErrorException(
                            \sprintf('Invalid character escape sequence: \\%s.', \chr($code))
                        );
                }

                ++$this->position;

                $chunkStart = $this->position;
            }
        }

        throw $this->createSyntaxErrorException('Unterminated string.');
    }

    /**
     * Reads a block string token from the source file.
     *
     * @param int   $line
     * @param int   $column
     * @param Token $prev
     * @return Token
     * @throws SyntaxErrorException
     */
    protected function lexBlockString(int $line, int $column, Token $prev): Token
    {
        $start          = $this->position;
        $this->position = $start + 3; // skip the triple-quote
        $chunkStart     = $this->position;
        $rawValue       = '';

        while ($this->position < $this->bodyLength && ($code = $this->readCharCode($this->position)) !== null) {
            // Closing Triple-Quote (""")
            if ($this->isTripleQuote($code)) {
                $rawValue .= sliceString($this->body, $chunkStart, $this->position);
                return new Token(
                    TokenKindEnum::BLOCK_STRING,
                    $start,
                    $this->position + 3,
                    $line,
                    $column,
                    $prev,
                    blockStringValue($rawValue)
                );
            }

            if (isSourceCharacter($code) && !isLineTerminator($code)) {
                throw $this->createSyntaxErrorException(
                    \sprintf('Invalid character within String: %s.', printCharCode($code))
                );
            }

            if ($this->isEscapedTripleQuote($code)) {
                $rawValue       .= sliceString($this->body, $chunkStart, $this->position) . '"""';
                $this->position += 4;
                $chunkStart     = $this->position;
            } else {
                ++$this->position;
            }
        }

        throw $this->createSyntaxErrorException('Unterminated string.');
    }

    /**
     * Skips whitespace at the current position.
     */
    protected function skipWhitespace(): void
    {
        while ($this->position < $this->bodyLength) {
            $code = $this->readCharCode($this->position);

            if (9 === $code || 32 === $code || 44 === $code || 0xfeff === $code) {
                // tab | space | comma | BOM
                ++$this->position;
            } elseif (10 === $code) {
                // new line (\n)
                ++$this->position;
                ++$this->line;
                $this->lineStart = $this->position;
            } elseif (13 === $code) {
                // carriage return (\r)
                if (10 === $this->readCharCode($this->position + 1)) {
                    // carriage return and new line (\r\n)
                    $this->position += 2;
                } else {
                    ++$this->position;
                }
                ++$this->line;
                $this->lineStart = $this->position;
            } else {
                break;
            }
        }
    }

    /**
     * @param int $position
     * @return int
     */
    protected function readCharCode(int $position): int
    {
        $char = \mb_substr($this->body, $position, 1, self::ENCODING);

        if ('' === $char) {
            return 0;
        }

        if (!isset(self::$charCodeCache[$char])) {
            $code = \ord($char);

            if ($code >= 128) {
                $code = \mb_ord($char, self::ENCODING);
            }

            self::$charCodeCache[$char] = $code;
        }

        return self::$charCodeCache[$char];
    }

    /**
     * Report a message that an unexpected character was encountered.
     *
     * @param int $code
     * @return string
     */
    protected function unexpectedCharacterMessage(int $code): string
    {
        if (isSourceCharacter($code) && !isLineTerminator($code)) {
            return \sprintf('Cannot contain the invalid character %s.', printCharCode($code));
        }

        if ($code === 39) {
            // '
            return 'Unexpected single quote character (\'), did you mean to use a double quote (")?';
        }

        return \sprintf('Cannot parse the unexpected character %s.', printCharCode($code));
    }

    /**
     * @param int $code
     * @return bool
     */
    protected function isSpread(int $code): bool
    {
        return 46 === $code &&
            $this->readCharCode($this->position + 1) === 46 &&
            $this->readCharCode($this->position + 2) === 46; // ...
    }

    /**
     * @param int $code
     * @return bool
     */
    protected function isString(int $code): bool
    {
        return 34 === $code && $this->readCharCode($this->position + 1) !== 34;
    }

    /**
     * @param int $code
     * @return bool
     */
    protected function isTripleQuote(int $code): bool
    {
        return 34 === $code &&
            34 === $this->readCharCode($this->position + 1) &&
            34 === $this->readCharCode($this->position + 2); // """
    }

    /**
     * @param int $code
     * @return bool
     */
    protected function isEscapedTripleQuote(int $code): bool
    {
        return $code === 92 &&
            34 === $this->readCharCode($this->position + 1) &&
            34 === $this->readCharCode($this->position + 2) &&
            34 === $this->readCharCode($this->position + 3); // \"""
    }
}


1			<?php
2
3			namespace Digia\GraphQL\Language;
4
5			use Digia\GraphQL\Error\SyntaxErrorException;
6
7			class Lexer implements LexerInterface
8			{
9			protected const ENCODING = 'UTF-8';
10
11			/**
12			* A map between punctuation character code and the corresponding token kind.
13			*
14			* @var array
15			*/
16			protected static $codeTokenKindMap = [
17			33 => TokenKindEnum::BANG,
18			36 => TokenKindEnum::DOLLAR,
19			38 => TokenKindEnum::AMP,
20			40 => TokenKindEnum::PAREN_L,
21			41 => TokenKindEnum::PAREN_R,
22			58 => TokenKindEnum::COLON,
23			61 => TokenKindEnum::EQUALS,
24			64 => TokenKindEnum::AT,
25			91 => TokenKindEnum::BRACKET_L,
26			93 => TokenKindEnum::BRACKET_R,
27			123 => TokenKindEnum::BRACE_L,
28			124 => TokenKindEnum::PIPE,
29			125 => TokenKindEnum::BRACE_R,
30			];
31
32			/**
33			* The source file for this lexer.
34			*
35			* @var Source
36			*/
37			protected $source;
38
39			/**
40			* The contents of the source file.
41			*
42			* @var string
43			*/
44			protected $body;
45
46			/**
47			* The total number of characters in the source file.
48			*
49			* @var int
50			*/
51			protected $bodyLength;
52
53			/**
54			* The options for this lexer.
55			*
56			* @var array
57			*/
58			protected $options = [];
59
60			/**
61			* The previously focused non-ignored token.
62			*
63			* @var Token
64			*/
65			protected $lastToken;
66
67			/**
68			* The currently focused non-ignored token.
69			*
70			* @var Token
71			*/
72			protected $token;
73
74			/**
75			* The current position.
76			*
77			* @var int
78			*/
79			protected $position;
80
81			/**
82			* The (1-indexed) line containing the current token.
83			*
84			* @var int
85			*/
86			protected $line;
87
88			/**
89			* The character offset at which the current line begins.
90			*
91			* @var int
92			*/
93			protected $lineStart;
94
95			/**
96			* A key-value map over characters and their corresponding character codes.
97			*
98			* @var array
99			*/
100			protected static $charCodeCache = [];
101
102			/**
103			* Lexer constructor.
104			* @param Source\|null $source
105			* @param array $options
106			*/
107			public function __construct(Source $source, array $options)
108			{
109			$startOfFileToken = $this->createStartOfFileToken();
110
111			$this->lastToken = $startOfFileToken;
112			$this->token = $startOfFileToken;
113			$this->line = 1;
114			$this->lineStart = 0;
115			$this->body = $source->getBody();
116			$this->bodyLength = \strlen($this->body);
117			$this->source = $source;
118			$this->options = $options;
119			}
120
121			/**
122			* @inheritdoc
123			* @throws SyntaxErrorException
124			*/
125			public function advance(): Token
126			{
127			$this->lastToken = $this->token;
128			return $this->token = $this->lookahead();
129			}
130
131			/**
132			* @inheritdoc
133			* @throws SyntaxErrorException
134			*/
135			public function lookahead(): Token
136			{
137			$token = $this->token;
138
139			if (TokenKindEnum::EOF !== $token->getKind()) {
140			do {
141			$next = $this->readToken($token);
142			$token->setNext($next);
143			$token = $next;
144			} while (TokenKindEnum::COMMENT === $token->getKind());
145			}
146
147			return $token;
148			}
149
150			/**
151			* @inheritdoc
152			*/
153			public function getOption(string $name, $default = null)
154			{
155			return $this->options[$name] ?? $default;
156			}
157
158			/**
159			* @inheritdoc
160			*/
161			public function getSource(): Source
162			{
163			return $this->source;
164			}
165
166			/**
167			* @inheritdoc
168			*/
169			public function getToken(): Token
170			{
171			return $this->token;
172			}
173
174			/**
175			* @inheritdoc
176			*/
177			public function getLastToken(): Token
178			{
179			return $this->lastToken;
180			}
181
182			/**
183			* @inheritdoc
184			*/
185			public function createSyntaxErrorException(?string $description = null): SyntaxErrorException
186			{
187			return new SyntaxErrorException(
188			$this->source,
189			$this->position,
190			$description ?? $this->unexpectedCharacterMessage($this->readCharCode($this->position))
191			);
192			}
193
194			/**
195			* Reads the token after the given token.
196			*
197			* @param Token $prev
198			* @return Token
199			* @throws SyntaxErrorException
200			*/
201			protected function readToken(Token $prev): Token
202			{
203			$this->position = $prev->getEnd();
204
205			$this->skipWhitespace();
206
207			$line = $this->line;
208			$column = (1 + $this->position) - $this->lineStart;
209
210			if ($this->position >= $this->bodyLength) {
211			return $this->createEndOfFileToken($line, $column, $prev);
212			}
213
214			$code = $this->readCharCode($this->position);
215
216			// Punctuation: [!$&:=@\|()\[\]{}]{1}
217			if (33 === $code \|\| 36 === $code \|\| 38 === $code \|\| 58 === $code \|\| 61 === $code \|\| 64 === $code \|\| 124 === $code \|\|
218			40 === $code \|\| 41 === $code \|\| 91 === $code \|\| 93 === $code \|\| 123 === $code \|\| 125 === $code) {
219			return $this->lexPunctuation($code, $line, $column, $prev);
220			}
221
222			// Comment: #[\u0009\u0020-\uFFFF]*
223			if (35 === $code) {
224			return $this->lexComment($line, $column, $prev);
225			}
226
227			// Int: -?(0\|[1-9][0-9]*)
228			// Float: -?(0\|[1-9][0-9]*)(\.[0-9]+)?((E\|e)(+\|-)?[0-9]+)?
229			if (45 === $code \|\| isNumber($code)) {
230			return $this->lexNumber($code, $line, $column, $prev);
231			}
232
233			// Name: [_A-Za-z][_0-9A-Za-z]*
234			if (isAlphaNumeric($code)) {
235			return $this->lexName($line, $column, $prev);
236			}
237
238			// Spread: ...
239			if ($this->bodyLength >= 3 && $this->isSpread($code)) {
240			return $this->lexSpread($line, $column, $prev);
241			}
242
243			// String: "([^"\\\u000A\u000D]\|(\\(u[0-9a-fA-F]{4}\|["\\/bfnrt])))*"
244			if ($this->isString($code)) {
245			return $this->lexString($line, $column, $prev);
246			}
247
248			// Block String: """("?"?(\\"""\|\\(?!=""")\|[^"\\]))*"""
249			if ($this->bodyLength >= 3 && $this->isTripleQuote($code)) {
250			return $this->lexBlockString($line, $column, $prev);
251			}
252
253			throw $this->createSyntaxErrorException();
254			}
255
256			/**
257			* @return Token
258			*/
259			protected function createStartOfFileToken(): Token
260			{
261			return new Token(TokenKindEnum::SOF);
262			}
263
264			/**
265			* Creates an End Of File (EOF) token.
266			*
267			* @param int $line
268			* @param int $column
269			* @param Token $prev
270			* @return Token
271			*/
272			protected function createEndOfFileToken(int $line, int $column, Token $prev): Token
273			{
274			return new Token(TokenKindEnum::EOF, $this->bodyLength, $this->bodyLength, $line, $column, $prev);
275			}
276
277			/**
278			* Reads a punctuation token from the source file.
279			*
280			* @param int $code
281			* @param int $line
282			* @param int $column
283			* @param Token $prev
284			* @return Token
285			* @throws SyntaxErrorException
286			*/
287			protected function lexPunctuation(int $code, int $line, int $column, Token $prev): ?Token
288			{
289			if (!isset(self::$codeTokenKindMap[$code])) {
290			throw $this->createSyntaxErrorException();
291			}
292
293			return new Token(self::$codeTokenKindMap[$code], $this->position, $this->position + 1, $line, $column, $prev);
294			}
295
296			/**
297			* Reads a name token from the source file.
298			*
299			* @param int $line
300			* @param int $column
301			* @param Token $prev
302			* @return Token
303			*/
304			protected function lexName(int $line, int $column, Token $prev): Token
305			{
306			$start = $this->position;
307
308			++$this->position;
309
310			while ($this->position !== $this->bodyLength &&
311			($code = $this->readCharCode($this->position)) !== null &&
312			isAlphaNumeric($code)) {
313			++$this->position;
314			}
315
316			$value = sliceString($this->body, $start, $this->position);
317
318			return new Token(TokenKindEnum::NAME, $start, $this->position, $line, $column, $prev, $value);
319			}
320
321			/**
322			* Reads a number (int or float) token from the source file.
323			*
324			* @param int $code
325			* @param int $line
326			* @param int $column
327			* @param Token $prev
328			* @return Token
329			* @throws SyntaxErrorException
330			*/
331			protected function lexNumber(int $code, int $line, int $column, Token $prev): Token
332			{
333			$start = $this->position;
334			$isFloat = false;
335
336			if (45 === $code) {
337			// -
338			$code = $this->readCharCode(++$this->position);
339			}
340
341			if (48 === $code) {
342			// 0
343			$code = $this->readCharCode(++$this->position);
344
345			if (isNumber($code)) {
346			throw $this->createSyntaxErrorException(
347			\sprintf('Invalid number, unexpected digit after 0: %s.', printCharCode($code))
348			);
349			}
350			} else {
351			$this->skipDigits($code);
352			$code = $this->readCharCode($this->position);
353			}
354
355			if (46 === $code) {
356			// .
357			$isFloat = true;
358
359			$code = $this->readCharCode(++$this->position);
360			$this->skipDigits($code);
361			$code = $this->readCharCode($this->position);
362			}
363
364			if (69 === $code \|\| 101 === $code) {
365			// e or E
366			$isFloat = true;
367
368			$code = $this->readCharCode(++$this->position);
369
370			if (43 === $code \|\| 45 === $code) {
371			// + or -
372			$code = $this->readCharCode(++$this->position);
373			}
374
375			$this->skipDigits($code);
376			}
377
378			return new Token(
379			$isFloat ? TokenKindEnum::FLOAT : TokenKindEnum::INT,
380			$start,
381			$this->position,
382			$line,
383			$column,
384			$prev,
385			sliceString($this->body, $start, $this->position)
386			);
387			}
388
389			/**
390			* Skips digits at the current position.
391			*
392			* @param int $code
393			* @throws SyntaxErrorException
394			*/
395			protected function skipDigits(int $code): void
396			{
397			if (isNumber($code)) {
398			do {
399			$code = $this->readCharCode(++$this->position);
400			} while (isNumber($code));
401
402			return;
403			}
404
405			throw $this->createSyntaxErrorException(
406			\sprintf('Invalid number, expected digit but got: %s.', printCharCode($code))
407			);
408			}
409
410			/**
411			* Reads a comment token from the source file.
412			*
413			* @param int $line
414			* @param int $column
415			* @param Token $prev
416			* @return Token
417			*/
418			protected function lexComment(int $line, int $column, Token $prev): Token
419			{
420			$start = $this->position;
421
422			do {
423			$code = $this->readCharCode(++$this->position);
424			} while ($code !== null && ($code > 0x001f \|\| 0x0009 === $code)); // SourceCharacter but not LineTerminator
425
426			return new Token(
427			TokenKindEnum::COMMENT,
428			$start,
429			$this->position,
430			$line,
431			$column,
432			$prev,
433			sliceString($this->body, $start + 1, $this->position)
434			);
435			}
436
437			/**
438			* Reads a spread token from the source.
439			*
440			* @param int $line
441			* @param int $column
442			* @param Token $prev
443			* @return Token
444			*/
445			protected function lexSpread(int $line, int $column, Token $prev): Token
446			{
447			return new Token(TokenKindEnum::SPREAD, $this->position, $this->position + 3, $line, $column, $prev);
448			}
449
450			/**
451			* Reads a string token from the source.
452			*
453			* @param int $line
454			* @param int $column
455			* @param Token $prev
456			* @return Token
457			* @throws SyntaxErrorException
458			*/
459			protected function lexString(int $line, int $column, Token $prev): Token
460			{
461			$start = $this->position;
462			$chunkStart = ++$this->position; // skip the quote
463			$value = '';
464
465			while ($this->position < $this->bodyLength &&
466			($code = $this->readCharCode($this->position)) !== null && !isLineTerminator($code)) {
467			// Closing Quote (")
468			if (34 === $code) {
469			$value .= sliceString($this->body, $chunkStart, $this->position);
470			return new Token(TokenKindEnum::STRING, $start, $this->position + 1, $line, $column, $prev, $value);
471			}
472
473			if (isSourceCharacter($code)) {
474			throw $this->createSyntaxErrorException(
475			\sprintf('Invalid character within String: %s.', printCharCode($code))
476			);
477			}
478
479			++$this->position;
480
481			if (92 === $code) {
482			// \
483			$value .= sliceString($this->body, $chunkStart, $this->position - 1);
484
485			$code = $this->readCharCode($this->position);
486
487			switch ($code) {
488			case 34: // "
489			$value .= '"';
490			break;
491			case 47: // /
492			$value .= '/';
493			break;
494			case 92: // \
495			$value .= '\\';
496			break;
497			case 98: // b
498			$value .= '\b';
499			break;
500			case 102: // f
501			$value .= '\f';
502			break;
503			case 110: // n
504			$value .= '\n';
505			break;
506			case 114: // r
507			$value .= '\r';
508			break;
509			case 116: // t
510			$value .= '\t';
511			break;
512			case 117: // u
513			$unicodeString = sliceString($this->body, $this->position + 1, $this->position + 5);
514
515			if (!\preg_match('/[0-9A-Fa-f]{4}/', $unicodeString)) {
516			throw $this->createSyntaxErrorException(
517			\sprintf('Invalid character escape sequence: \\u%s.', $unicodeString)
518			);
519			}
520
521			$value .= '\\u' . $unicodeString;
522
523			$this->position += 4;
524
525			break;
526			default:
527			throw $this->createSyntaxErrorException(
528			\sprintf('Invalid character escape sequence: \\%s.', \chr($code))
529			);
530			}
531
532			++$this->position;
533
534			$chunkStart = $this->position;
535			}
536			}
537
538			throw $this->createSyntaxErrorException('Unterminated string.');
539			}
540
541			/**
542			* Reads a block string token from the source file.
543			*
544			* @param int $line
545			* @param int $column
546			* @param Token $prev
547			* @return Token
548			* @throws SyntaxErrorException
549			*/
550			protected function lexBlockString(int $line, int $column, Token $prev): Token
551			{
552			$start = $this->position;
553			$this->position = $start + 3; // skip the triple-quote
554			$chunkStart = $this->position;
555			$rawValue = '';
556
557			while ($this->position < $this->bodyLength && ($code = $this->readCharCode($this->position)) !== null) {
558			// Closing Triple-Quote (""")
559			if ($this->isTripleQuote($code)) {
560			$rawValue .= sliceString($this->body, $chunkStart, $this->position);
561			return new Token(
562			TokenKindEnum::BLOCK_STRING,
563			$start,
564			$this->position + 3,
565			$line,
566			$column,
567			$prev,
568			blockStringValue($rawValue)
569			);
570			}
571
572			if (isSourceCharacter($code) && !isLineTerminator($code)) {
573			throw $this->createSyntaxErrorException(
574			\sprintf('Invalid character within String: %s.', printCharCode($code))
575			);
576			}
577
578			if ($this->isEscapedTripleQuote($code)) {
579			$rawValue .= sliceString($this->body, $chunkStart, $this->position) . '"""';
580			$this->position += 4;
581			$chunkStart = $this->position;
582			} else {
583			++$this->position;
584			}
585			}
586
587			throw $this->createSyntaxErrorException('Unterminated string.');
588			}
589
590			/**
591			* Skips whitespace at the current position.
592			*/
593			protected function skipWhitespace(): void
594			{
595			while ($this->position < $this->bodyLength) {
596			$code = $this->readCharCode($this->position);
597
598			if (9 === $code \|\| 32 === $code \|\| 44 === $code \|\| 0xfeff === $code) {
599			// tab \| space \| comma \| BOM
600			++$this->position;
601			} elseif (10 === $code) {
602			// new line (\n)
603			++$this->position;
604			++$this->line;
605			$this->lineStart = $this->position;
606			} elseif (13 === $code) {
607			// carriage return (\r)
608			if (10 === $this->readCharCode($this->position + 1)) {
609			// carriage return and new line (\r\n)
610			$this->position += 2;
611			} else {
612			++$this->position;
613			}
614			++$this->line;
615			$this->lineStart = $this->position;
616			} else {
617			break;
618			}
619			}
620			}
621
622			/**
623			* @param int $position
624			* @return int
625			*/
626			protected function readCharCode(int $position): int
627			{
628			$char = \mb_substr($this->body, $position, 1, self::ENCODING);
629
630			if ('' === $char) {
631			return 0;
632			}
633
634			if (!isset(self::$charCodeCache[$char])) {
635			$code = \ord($char);
636
637			if ($code >= 128) {
638			$code = \mb_ord($char, self::ENCODING);
639			}
640
641			self::$charCodeCache[$char] = $code;
642			}
643
644			return self::$charCodeCache[$char];
645			}
646
647			/**
648			* Report a message that an unexpected character was encountered.
649			*
650			* @param int $code
651			* @return string
652			*/
653			protected function unexpectedCharacterMessage(int $code): string
654			{
655			if (isSourceCharacter($code) && !isLineTerminator($code)) {
656			return \sprintf('Cannot contain the invalid character %s.', printCharCode($code));
657			}
658
659			if ($code === 39) {
660			// '
661			return 'Unexpected single quote character (\'), did you mean to use a double quote (")?';
662			}
663
664			return \sprintf('Cannot parse the unexpected character %s.', printCharCode($code));
665			}
666
667			/**
668			* @param int $code
669			* @return bool
670			*/
671			protected function isSpread(int $code): bool
672			{
673			return 46 === $code &&
674			$this->readCharCode($this->position + 1) === 46 &&
675			$this->readCharCode($this->position + 2) === 46; // ...
676			}
677
678			/**
679			* @param int $code
680			* @return bool
681			*/
682			protected function isString(int $code): bool
683			{
684			return 34 === $code && $this->readCharCode($this->position + 1) !== 34;
685			}
686
687			/**
688			* @param int $code
689			* @return bool
690			*/
691			protected function isTripleQuote(int $code): bool
692			{
693			return 34 === $code &&
694			34 === $this->readCharCode($this->position + 1) &&
695			34 === $this->readCharCode($this->position + 2); // """
696			}
697
698			/**
699			* @param int $code
700			* @return bool
701			*/
702			protected function isEscapedTripleQuote(int $code): bool
703			{
704			return $code === 92 &&
705			34 === $this->readCharCode($this->position + 1) &&
706			34 === $this->readCharCode($this->position + 2) &&
707			34 === $this->readCharCode($this->position + 3); // \"""
708			}
709			}
710

digiaonline / graphql-php

Pull Request — master (#196)

Lexer::readToken() C

Complexity

Size

Duplication

Importance

How to fix Long Method Complexity

Long Method

Duplication Side-by-Side

Filter issues like