Json5Decoder::next() - Code Metrics - Inspection of "Optimizations" - colinodell/json5 - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#2)

by Colin

created 2018-01-14 13:53 UTC

Json5Decoder::next() A

↳ Parent: Json5Decoder

Complexity

Conditions	4
Paths	2

Size

Total Lines	17
Code Lines	10

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	10
CRAP Score	4

Importance

Changes	2
Bugs	0	Features	0

Metric	Value
cc	4
eloc	10
c	2
b	0
f	0
nc	2
nop	0
dl	0
loc	17
ccs	10
cts	10
cp	1
crap	4
rs	9.2

<?php

/*
 * This file is part of the colinodell/json5 package.
 *
 * (c) Colin O'Dell <[email protected]>
 *
 * Based on the official JSON5 implementation for JavaScript (https://github.com/json5/json5)
 *  - (c) 2012-2016 Aseem Kishore and others (https://github.com/json5/json5/contributors)
 *
 * For the full copyright and license information, please view the LICENSE
 * file that was distributed with this source code.
 */

namespace ColinODell\Json5;

final class Json5Decoder
{
    private $at = 0;

    private $lineNumber = 1;

    private $columnNumber = 1;

    private $ch;

    private $chArr;

    private $associative = false;

    private $maxDepth = 512;

    private $castBigIntToString = false;

    private $depth = 1;

    private $length;

    private $remainderCache;

    private $remainderCacheAt;

    /**
     * Private constructor.
     *
     * @param string $json
     * @param bool   $associative
     * @param int    $depth
     * @param bool   $castBigIntToString
     */
    private function __construct($json, $associative = false, $depth = 512, $castBigIntToString = false)
    {
        $this->associative = $associative;
        $this->maxDepth = $depth;
        $this->castBigIntToString = $castBigIntToString;

        $this->length = mb_strlen($json, 'utf-8');

        $this->chArr = preg_split('//u', $json, null, PREG_SPLIT_NO_EMPTY);
        $this->ch = $this->charAt(0);

        $this->remainderCache = $json;
        $this->remainderCacheAt = 0;
    }

    /**
     * Takes a JSON encoded string and converts it into a PHP variable.
     *
     * The parameters exactly match PHP's json_decode() function - see
     * http://php.net/manual/en/function.json-decode.php for more information.
     *
     * @param string $source      The JSON string being decoded.
     * @param bool   $associative When TRUE, returned objects will be converted into associative arrays.
     * @param int    $depth       User specified recursion depth.
     * @param int    $options     Bitmask of JSON decode options.
     *
     * @return mixed
     */
    public static function decode($source, $associative = false, $depth = 512, $options = 0)
    {
        // Try parsing with json_decode first, since that's much faster
        $result = json_decode($source, $associative, $depth, $options);
        if (json_last_error() === JSON_ERROR_NONE) {
            return $result;
        }

        // Fall back to JSON5 if that fails
        $associative = $associative === true || ($associative === null && $options & JSON_OBJECT_AS_ARRAY);
        $castBigIntToString = $options & JSON_BIGINT_AS_STRING;

        $decoder = new self((string)$source, $associative, $depth, $castBigIntToString);
function acceptsInteger($int) { }

$x = '123'; // string "123"

// Instead of
acceptsInteger($x);

// we recommend to use
acceptsInteger((integer) $x);

        $result = $decoder->value();
        $decoder->white();
        if ($decoder->ch) {
            $decoder->throwSyntaxError('Syntax error');
        }

        return $result;
    }

    /**
     * @param int $at
     *
     * @return string|null
     */
    private function charAt($at)
    {
        if ($at >= $this->length) {
            return null;
        }

        return $this->chArr[$at];
    }

    /**
     * Parse the next character.
     *
     * @return null|string
     */
    private function next()
    {
        // Get the next character. When there are no more characters,
        // return the empty string.
        if ($this->ch === "\n" || ($this->ch === "\r" && $this->peek() !== "\n")) {
            $this->at++;
            $this->lineNumber++;
            $this->columnNumber = 1;
        } else {
            $this->at++;
            $this->columnNumber++;
        }

        $this->ch = $this->charAt($this->at);

        return $this->ch;
    }

    /**
     * Parse the next character if it matches $c or fail.
     *
     * @param string $c
     *
     * @return string|null
     */
    private function nextOrFail($c)
    {
        if ($c !== $this->ch) {
            $this->throwSyntaxError(sprintf(
                'Expected %s instead of %s',
                self::renderChar($c),
                self::renderChar($this->ch)
            ));
        }

        return $this->next();
    }

    /**
     * Get the next character without consuming it or
     * assigning it to the ch variable.
     *
     * @return mixed
     */
    private function peek()
    {
        return $this->charAt($this->at + 1);
    }

    /**
     * Attempt to match a regular expression at the current position on the current line.
     *
     * This function will not match across multiple lines.
     *
     * @param string $regex
     *
     * @return string|null
     */
    private function match($regex)
    {
        $subject = $this->getRemainder();

        $matches = [];
        if (!preg_match($regex, $subject, $matches, PREG_OFFSET_CAPTURE)) {
            return null;
        }

        // PREG_OFFSET_CAPTURE always returns the byte offset, not the char offset, which is annoying
        $offset = mb_strlen(mb_strcut($subject, 0, $matches[0][1], 'utf-8'), 'utf-8');

        // [0][0] contains the matched text

        // [0][1] contains the index of that match
        $advanceBy = $offset + mb_strlen($matches[0][0], 'utf-8');

        $this->at += $advanceBy;
        $this->columnNumber += $advanceBy;
        $this->ch = $this->charAt($this->at);

        return $matches[0][0];
    }

    /**
     * Parse an identifier.
     *
     * Normally, reserved words are disallowed here, but we
     * only use this for unquoted object keys, where reserved words are allowed,
     * so we don't check for those here. References:
     * - http://es5.github.com/#x7.6
     * - https://developer.mozilla.org/en/Core_JavaScript_1.5_Guide/Core_Language_Features#Variables
     * - http://docstore.mik.ua/orelly/webprog/jscript/ch02_07.htm
     */
    private function identifier()
    {
        // @codingStandardsIgnoreStart
        // Be careful when editing this regex, there are a couple Unicode characters in between here -------------vv
        $match = $this->match('/^(?:[\$_\p{L}\p{Nl}]|\\\\u[0-9A-Fa-f]{4})(?:[\$_\p{L}\p{Nl}\p{Mn}\p{Mc}\p{Nd}\p{Pc}‌‍]|\\\\u[0-9A-Fa-f]{4})*/u');
        // @codingStandardsIgnoreEnd

        if ($match === null) {
            $this->throwSyntaxError('Bad identifier as unquoted key');
        }

        // Un-escape escaped Unicode chars
        $unescaped = preg_replace_callback('/(?:\\\\u[0-9A-Fa-f]{4})+/', function ($m) {
            return json_decode('"'.$m[0].'"');
        }, $match);

        return $unescaped;
    }

    private function number()
    {
        $number = null;
        $sign = '';
        $string = '';
        $base = 10;

        if ($this->ch === '-' || $this->ch === '+') {
            $sign = $this->ch;
            $this->next($this->ch);

        }

        // support for Infinity
        if ($this->ch === 'I') {
            $number = $this->word();
            if ($number === null) {
                $this->throwSyntaxError('Unexpected word for number');
            }

            return ($sign === '-') ? -INF : INF;
        }

        // support for NaN
        if ($this->ch === 'N') {
            $number = $this->word();
            if ($number !== NAN) {
                $this->throwSyntaxError('expected word to be NaN');
            }

            // ignore sign as -NaN also is NaN
            return $number;
        }

        if ($this->ch === '0') {
            $string .= $this->ch;
            $this->next();
            if ($this->ch === 'x' || $this->ch === 'X') {
                $string .= $this->ch;
                $this->next();
                $base = 16;
            } elseif (is_numeric($this->ch)) {
                $this->throwSyntaxError('Octal literal');
            }
        }

        switch ($base) {
            case 10:
                if ((is_numeric($this->ch) || $this->ch === '.') && ($match = $this->match('/^\d*\.?\d*/')) !== null) {

                    $string .= $match;
                }
                if (($this->ch === 'E' || $this->ch === 'e') && ($match = $this->match('/^[Ee][-+]?\d*/')) !== null) {

                    $string .= $match;
                }
                $number = $string;
                break;
            case 16:
                if (($match = $this->match('/^[A-Fa-f0-9]+/')) !== null) {
                    $string .= $match;
                    $number = hexdec($string);
                    break;
                }
                $this->throwSyntaxError('Bad hex number');
        }

        if ($sign === '-') {
            $number = -$number;
        }

        if (!is_numeric($number) || !is_finite($number)) {
            $this->throwSyntaxError('Bad number');
        }

        if ($this->castBigIntToString) {
            return $number;
        }

        // Adding 0 will automatically cast this to an int or float
        return $number + 0;
    }

    private function string()
    {
        $string = '';

        $delim = $this->ch;
        $this->next();
        while ($this->ch !== null) {
            if ($this->ch === $delim) {
                $this->next();

                return $string;
            }

            if ($this->ch === '\\') {
                if ($this->peek() === 'u' && $unicodeEscaped = $this->match('/^(?:\\\\u[A-Fa-f0-9]{4})+/')) {
                    $string .= json_decode('"'.$unicodeEscaped.'"');
                    continue;
                }

                $this->next();
                if ($this->ch === "\r") {
                    if ($this->peek() === "\n") {
                        $this->next();
                    }
                } elseif (($escapee = self::getEscapee($this->ch)) !== null) {
                    $string .= $escapee;
                } else {
                    break;
                }
            } elseif ($this->ch === "\n") {
                // unescaped newlines are invalid; see:
                // https://github.com/json5/json5/issues/24
                // @todo this feels special-cased; are there other invalid unescaped chars?
                break;
            } else {
                $string .= $this->ch;
            }

            $this->next();
        }

        $this->throwSyntaxError('Bad string');
    }

    /**
     * Skip an inline comment, assuming this is one.
     *
     * The current character should be the second / character in the // pair that begins this inline comment.
     * To finish the inline comment, we look for a newline or the end of the text.
     */
    private function inlineComment()
    {
        do {
            $this->next();
            if ($this->ch === "\n" || $this->ch === "\r") {
                $this->next();

                return;
            }
        } while ($this->ch !== null);
    }

    /**
     * Skip a block comment, assuming this is one.
     *
     * The current character should be the * character in the /* pair that begins this block comment.
     * To finish the block comment, we look for an ending *​/ pair of characters,
     * but we also watch for the end of text before the comment is terminated.
     */
    private function blockComment()
    {
        do {
            $this->next();
            while ($this->ch === '*') {
                $this->nextOrFail('*');
                if ($this->ch === '/') {
                    $this->nextOrFail('/');

                    return;
                }
            }
        } while ($this->ch !== null);

        $this->throwSyntaxError('Unterminated block comment');
    }

    /**
     * Skip a comment, whether inline or block-level, assuming this is one.
     */
    private function comment()
    {
        // Comments always begin with a / character.
        $this->nextOrFail('/');

        if ($this->ch === '/') {
            $this->inlineComment();
        } elseif ($this->ch === '*') {
            $this->blockComment();
        } else {
            $this->throwSyntaxError('Unrecognized comment');
        }
    }

    /**
     * Skip whitespace and comments.
     *
     * Note that we're detecting comments by only a single / character.
     * This works since regular expressions are not valid JSON(5), but this will
     * break if there are other valid values that begin with a / character!
     */
    private function white()
    {
        while ($this->ch !== null) {
            if ($this->ch === '/') {
                $this->comment();
            } elseif (preg_match('/[ \t\r\n\v\f\xA0\x{FEFF}]/u', $this->ch) === 1) {
                $this->next();
            } else {
                return;
            }
        }
    }

    /**
     * Matches true, false, null, etc
     */
    private function word()
    {
        switch ($this->ch) {
            case 't':
                $this->nextOrFail('t');
                $this->nextOrFail('r');
                $this->nextOrFail('u');
                $this->nextOrFail('e');
                return true;
            case 'f':
                $this->nextOrFail('f');
                $this->nextOrFail('a');
                $this->nextOrFail('l');
                $this->nextOrFail('s');
                $this->nextOrFail('e');
                return false;
            case 'n':
                $this->nextOrFail('n');
                $this->nextOrFail('u');
                $this->nextOrFail('l');
                $this->nextOrFail('l');
                return null;
            case 'I':
                $this->nextOrFail('I');
                $this->nextOrFail('n');
                $this->nextOrFail('f');
                $this->nextOrFail('i');
                $this->nextOrFail('n');
                $this->nextOrFail('i');
                $this->nextOrFail('t');
                $this->nextOrFail('y');
                return INF;
            case 'N':
                $this->nextOrFail('N');
                $this->nextOrFail('a');
                $this->nextOrFail('N');
                return NAN;
        }

        $this->throwSyntaxError('Unexpected ' . self::renderChar($this->ch));
    }

    private function arr()
    {
        $arr = [];

        if ($this->ch === '[') {
            if (++$this->depth > $this->maxDepth) {
                $this->throwSyntaxError('Maximum stack depth exceeded');
            }

            $this->nextOrFail('[');
            $this->white();
            while ($this->ch !== null) {
                if ($this->ch === ']') {
                    $this->nextOrFail(']');
                    $this->depth--;
                    return $arr; // Potentially empty array
                }
                // ES5 allows omitting elements in arrays, e.g. [,] and
                // [,null]. We don't allow this in JSON5.
                if ($this->ch === ',') {
                    $this->throwSyntaxError('Missing array element');
                }

                $arr[] = $this->value();

                $this->white();
                // If there's no comma after this value, this needs to
                // be the end of the array.
                if ($this->ch !== ',') {
                    $this->nextOrFail(']');
                    $this->depth--;
                    return $arr;
                }
                $this->nextOrFail(',');
                $this->white();
            }
        }

        $this->throwSyntaxError('Bad array');
    }

    /**
     * Parse an object value
     */
    private function obj()
    {
        $object = $this->associative ? [] : new \stdClass;

        if ($this->ch === '{') {
            if (++$this->depth > $this->maxDepth) {
                $this->throwSyntaxError('Maximum stack depth exceeded');
            }

            $this->nextOrFail('{');
            $this->white();
            while ($this->ch) {
                if ($this->ch === '}') {
                    $this->nextOrFail('}');
                    $this->depth--;
                    return $object; // Potentially empty object
                }

                // Keys can be unquoted. If they are, they need to be
                // valid JS identifiers.
                if ($this->ch === '"' || $this->ch === "'") {
                    $key = $this->string();
                } else {
                    $key = $this->identifier();
                }

                $this->white();
                $this->nextOrFail(':');
                if ($this->associative) {
                    $object[$key] = $this->value();
                } else {
                    $object->{$key} = $this->value();
                }
                $this->white();
                // If there's no comma after this pair, this needs to be
                // the end of the object.
                if ($this->ch !== ',') {
                    $this->nextOrFail('}');
                    $this->depth--;
                    return $object;
                }
                $this->nextOrFail(',');
                $this->white();
            }
        }

        $this->throwSyntaxError('Bad object');
    }

    /**
     * Parse a JSON value.
     *
     * It could be an object, an array, a string, a number,
     * or a word.
     */
    private function value()
    {
        $this->white();
        switch ($this->ch) {
            case '{':
                return $this->obj();
            case '[':
                return $this->arr();
            case '"':
            case "'":
                return $this->string();
            case '-':
            case '+':
            case '.':
                return $this->number();
            default:
                return is_numeric($this->ch) ? $this->number() : $this->word();
        }
    }

    private function throwSyntaxError($message)
    {
        throw new SyntaxError($message, $this->lineNumber, $this->columnNumber);
    }

    private static function renderChar($chr)
    {
        return $chr === null ? 'EOF' : "'" . $chr . "'";
    }

    /**
     * @param string $ch
     *
     * @return string|null
     */
    private static function getEscapee($ch)
    {
        switch ($ch) {
            // @codingStandardsIgnoreStart
            case "'":  return "'";
            case '"':  return '"';
            case '\\': return '\\';
            case '/':  return '/';
            case "\n": return '';
            case 'b':  return chr(8);
            case 'f':  return "\f";
            case 'n':  return "\n";
            case 'r':  return "\r";
            case 't':  return "\t";
            default:   return null;
            // @codingStandardsIgnoreEnd
        }
    }

    /**
     * Returns everything from $this->at onwards.
     *
     * Utilizes a cache so we don't have to continuously parse through UTF-8
     * data that was earlier in the string which we don't even care about.
     *
     * @return string
     */
    private function getRemainder()
    {
        if ($this->remainderCacheAt === $this->at) {
            return $this->remainderCache;
        }

        $subject = mb_substr($this->remainderCache, $this->at - $this->remainderCacheAt);
        $this->remainderCache = $subject;
        $this->remainderCacheAt = $this->at;

        return $subject;
    }
}


1			<?php
2
3			/*
4			* This file is part of the colinodell/json5 package.
5			*
6			* (c) Colin O'Dell <[email protected]>
7			*
8			* Based on the official JSON5 implementation for JavaScript (https://github.com/json5/json5)
9			* - (c) 2012-2016 Aseem Kishore and others (https://github.com/json5/json5/contributors)
10			*
11			* For the full copyright and license information, please view the LICENSE
12			* file that was distributed with this source code.
13			*/
14
15			namespace ColinODell\Json5;
16
17			final class Json5Decoder
18			{
19			private $at = 0;
20
21			private $lineNumber = 1;
22
23			private $columnNumber = 1;
24
25			private $ch;
26
27			private $chArr;
28
29			private $associative = false;
30
31			private $maxDepth = 512;
32
33			private $castBigIntToString = false;
34
35			private $depth = 1;
36
37			private $length;
38
39			private $remainderCache;
40
41			private $remainderCacheAt;
42
43			/**
44			* Private constructor.
45			*
46			* @param string $json
47			* @param bool $associative
48			* @param int $depth
49			* @param bool $castBigIntToString
50			*/
51	223		private function __construct($json, $associative = false, $depth = 512, $castBigIntToString = false)
52			{
53	223		$this->associative = $associative;
54	223		$this->maxDepth = $depth;
55	223		$this->castBigIntToString = $castBigIntToString;
56
57	223		$this->length = mb_strlen($json, 'utf-8');
58
59	223		$this->chArr = preg_split('//u', $json, null, PREG_SPLIT_NO_EMPTY);
60	223		$this->ch = $this->charAt(0);
61
62	223		$this->remainderCache = $json;
63	223		$this->remainderCacheAt = 0;
64	223		}
65
66			/**
67			* Takes a JSON encoded string and converts it into a PHP variable.
68			*
69			* The parameters exactly match PHP's json_decode() function - see
70			* http://php.net/manual/en/function.json-decode.php for more information.
71			*
72			* @param string $source The JSON string being decoded.
73			* @param bool $associative When TRUE, returned objects will be converted into associative arrays.
74			* @param int $depth User specified recursion depth.
75			* @param int $options Bitmask of JSON decode options.
76			*
77			* @return mixed
78			*/
79	396		public static function decode($source, $associative = false, $depth = 512, $options = 0)
80			{
81			// Try parsing with json_decode first, since that's much faster
82	396		$result = json_decode($source, $associative, $depth, $options);
83	396		if (json_last_error() === JSON_ERROR_NONE) {
84	173		return $result;
85			}
86
87			// Fall back to JSON5 if that fails
88	223		$associative = $associative === true \|\| ($associative === null && $options & JSON_OBJECT_AS_ARRAY);
89	223		$castBigIntToString = $options & JSON_BIGINT_AS_STRING;
90
91	223		$decoder = new self((string)$source, $associative, $depth, $castBigIntToString);
			0 ignored issues – show Documentation introduced 2017-11-11 13:06 UTC by Report Bug Copy Issue Report `$castBigIntToString` is of type `integer`, but the function expects a `boolean`. It seems like the type of the argument is not accepted by the function/method which you are calling. In some cases, in particular if PHP’s automatic type-juggling kicks in this might be fine. In other cases, however this might be a bug. We suggest to add an explicit type cast like in the following example: function acceptsInteger($int) { } $x = '123'; // string "123" // Instead of acceptsInteger($x); // we recommend to use acceptsInteger((integer) $x); Loading history...
92
93	223		$result = $decoder->value();
94	152		$decoder->white();
95	149		if ($decoder->ch) {
96	18		$decoder->throwSyntaxError('Syntax error');
97			}
98
99	131		return $result;
100			}
101
102			/**
103			* @param int $at
104			*
105			* @return string\|null
106			*/
107	223		private function charAt($at)
108			{
109	223		if ($at >= $this->length) {
110	144		return null;
111			}
112
113	222		return $this->chArr[$at];
114			}
115
116			/**
117			* Parse the next character.
118			*
119			* @return null\|string
120			*/
121	201		private function next()
122			{
123			// Get the next character. When there are no more characters,
124			// return the empty string.
125	201		if ($this->ch === "\n" \|\| ($this->ch === "\r" && $this->peek() !== "\n")) {
126	164		$this->at++;
127	164		$this->lineNumber++;
128	164		$this->columnNumber = 1;
129	94		} else {
130	197		$this->at++;
131	197		$this->columnNumber++;
132			}
133
134	201		$this->ch = $this->charAt($this->at);
135
136	201		return $this->ch;
137			}
138
139			/**
140			* Parse the next character if it matches $c or fail.
141			*
142			* @param string $c
143			*
144			* @return string\|null
145			*/
146	138		private function nextOrFail($c)
147			{
148	138		if ($c !== $this->ch) {
149	18		$this->throwSyntaxError(sprintf(
150	18		'Expected %s instead of %s',
151	18		self::renderChar($c),
152	18		self::renderChar($this->ch)
153	12		));
154			}
155
156	138		return $this->next();
157			}
158
159			/**
160			* Get the next character without consuming it or
161			* assigning it to the ch variable.
162			*
163			* @return mixed
164			*/
165	30		private function peek()
166			{
167	30		return $this->charAt($this->at + 1);
168			}
169
170			/**
171			* Attempt to match a regular expression at the current position on the current line.
172			*
173			* This function will not match across multiple lines.
174			*
175			* @param string $regex
176			*
177			* @return string\|null
178			*/
179	103		private function match($regex)
180			{
181	103		$subject = $this->getRemainder();
182
183	103		$matches = [];
184	103		if (!preg_match($regex, $subject, $matches, PREG_OFFSET_CAPTURE)) {
185	15		return null;
186			}
187
188			// PREG_OFFSET_CAPTURE always returns the byte offset, not the char offset, which is annoying
189	88		$offset = mb_strlen(mb_strcut($subject, 0, $matches[0][1], 'utf-8'), 'utf-8');
190
191			// [0][0] contains the matched text
			0 ignored issues – show Unused Code Comprehensibility introduced 2017-11-10 22:03 UTC by Report Bug Copy Issue Report `40%` of this comment could be valid code. Did you maybe forget this after debugging? Sometimes obsolete code just ends up commented out instead of removed. In this case it is better to remove the code once you have checked you do not need it. The code might also have been commented out for debugging purposes. In this case it is vital that someone uncomments it again or your project may behave in very unexpected ways in production. This check looks for comments that seem to be mostly valid code and reports them. Loading history...
192			// [0][1] contains the index of that match
193	88		$advanceBy = $offset + mb_strlen($matches[0][0], 'utf-8');
194
195	88		$this->at += $advanceBy;
196	88		$this->columnNumber += $advanceBy;
197	88		$this->ch = $this->charAt($this->at);
198
199	88		return $matches[0][0];
200			}
201
202			/**
203			* Parse an identifier.
204			*
205			* Normally, reserved words are disallowed here, but we
206			* only use this for unquoted object keys, where reserved words are allowed,
207			* so we don't check for those here. References:
208			* - http://es5.github.com/#x7.6
209			* - https://developer.mozilla.org/en/Core_JavaScript_1.5_Guide/Core_Language_Features#Variables
210			* - http://docstore.mik.ua/orelly/webprog/jscript/ch02_07.htm
211			*/
212	42		private function identifier()
213			{
214			// @codingStandardsIgnoreStart
215			// Be careful when editing this regex, there are a couple Unicode characters in between here -------------vv
216	42		$match = $this->match('/^(?:[\$_\p{L}\p{Nl}]\|\\\\u[0-9A-Fa-f]{4})(?:[\$_\p{L}\p{Nl}\p{Mn}\p{Mc}\p{Nd}\p{Pc}‌‍]\|\\\\u[0-9A-Fa-f]{4})*/u');
217			// @codingStandardsIgnoreEnd
218
219	42		if ($match === null) {
220	9		$this->throwSyntaxError('Bad identifier as unquoted key');
221			}
222
223			// Un-escape escaped Unicode chars
224	33		$unescaped = preg_replace_callback('/(?:\\\\u[0-9A-Fa-f]{4})+/', function ($m) {
225	6		return json_decode('"'.$m[0].'"');
226	33		}, $match);
227
228	33		return $unescaped;
229			}
230
231	78		private function number()
232			{
233	78		$number = null;
234	78		$sign = '';
235	78		$string = '';
236	78		$base = 10;
237
238	78		if ($this->ch === '-' \|\| $this->ch === '+') {
239	40		$sign = $this->ch;
240	40		$this->next($this->ch);
			0 ignored issues – show Unused Code introduced 2018-01-13 23:13 UTC by Report Bug Copy Issue Report The call to `Json5Decoder::next()` has too many arguments starting with `$this->ch`. This check compares calls to functions or methods with their respective definitions. If the call has more arguments than are defined, it raises an issue. If a function is defined several times with a different number of parameters, the check may pick up the wrong definition and report false positives. One codebase where this has been known to happen is Wordpress. In this case you can add the `@ignore` PhpDoc annotation to the duplicate definition and it will be ignored. Loading history...
241	14		}
242
243			// support for Infinity
244	78		if ($this->ch === 'I') {
245	6		$number = $this->word();
246	6		if ($number === null) {
247			$this->throwSyntaxError('Unexpected word for number');
248			}
249
250	6		return ($sign === '-') ? -INF : INF;
251			}
252
253			// support for NaN
254	72		if ($this->ch === 'N') {
255			$number = $this->word();
256			if ($number !== NAN) {
257			$this->throwSyntaxError('expected word to be NaN');
258			}
259
260			// ignore sign as -NaN also is NaN
261			return $number;
262			}
263
264	72		if ($this->ch === '0') {
265	39		$string .= $this->ch;
266	39		$this->next();
267	39		if ($this->ch === 'x' \|\| $this->ch === 'X') {
268	23		$string .= $this->ch;
269	23		$this->next();
270	23		$base = 16;
271	28		} elseif (is_numeric($this->ch)) {
272	10		$this->throwSyntaxError('Octal literal');
273			}
274	12		}
275
276			switch ($base) {
277	62		case 10:
278	42	View Code Duplication	if ((is_numeric($this->ch) \|\| $this->ch === '.') && ($match = $this->match('/^\d\.?\d/')) !== null) {
			0 ignored issues – show Duplication introduced 2018-01-13 23:13 UTC by Report Bug Copy Issue Report This code seems to be duplicated across your project. Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation. You can also find more detailed suggestions in the “Code” section of your repository. Loading history...
279	41		$string .= $match;
280	16		}
281	42	View Code Duplication	if (($this->ch === 'E' \|\| $this->ch === 'e') && ($match = $this->match('/^[Ee][-+]?\d*/')) !== null) {
			0 ignored issues – show Duplication introduced 2018-01-13 23:13 UTC by Report Bug Copy Issue Report This code seems to be duplicated across your project. Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation. You can also find more detailed suggestions in the “Code” section of your repository. Loading history...
282	19		$string .= $match;
283	12		}
284	42		$number = $string;
285	42		break;
286	23		case 16:
287	23		if (($match = $this->match('/^[A-Fa-f0-9]+/')) !== null) {
288	20		$string .= $match;
289	20		$number = hexdec($string);
290	20		break;
291			}
292	3		$this->throwSyntaxError('Bad hex number');
293			}
294
295	59		if ($sign === '-') {
296	10		$number = -$number;
297	4		}
298
299	59		if (!is_numeric($number) \|\| !is_finite($number)) {
300	3		$this->throwSyntaxError('Bad number');
301			}
302
303	56		if ($this->castBigIntToString) {
304			return $number;
305			}
306
307			// Adding 0 will automatically cast this to an int or float
308	56		return $number + 0;
309			}
310
311	63		private function string()
312			{
313	63		$string = '';
314
315	63		$delim = $this->ch;
316	63		$this->next();
317	63		while ($this->ch !== null) {
318	63		if ($this->ch === $delim) {
319	57		$this->next();
320
321	57		return $string;
322			}
323
324	63		if ($this->ch === '\\') {
325	24		if ($this->peek() === 'u' && $unicodeEscaped = $this->match('/^(?:\\\\u[A-Fa-f0-9]{4})+/')) {
326			$string .= json_decode('"'.$unicodeEscaped.'"');
327			continue;
328			}
329
330	24		$this->next();
331	24		if ($this->ch === "\r") {
332	6		if ($this->peek() === "\n") {
333	4		$this->next();
334	2		}
335	22		} elseif (($escapee = self::getEscapee($this->ch)) !== null) {
336	15		$string .= $escapee;
337	10		} else {
338	10		break;
339			}
340	63		} elseif ($this->ch === "\n") {
341			// unescaped newlines are invalid; see:
342			// https://github.com/json5/json5/issues/24
343			// @todo this feels special-cased; are there other invalid unescaped chars?
344	3		break;
345			} else {
346	63		$string .= $this->ch;
347			}
348
349	63		$this->next();
350	42		}
351
352	6		$this->throwSyntaxError('Bad string');
353			}
354
355			/**
356			* Skip an inline comment, assuming this is one.
357			*
358			* The current character should be the second / character in the // pair that begins this inline comment.
359			* To finish the inline comment, we look for a newline or the end of the text.
360			*/
361	36		private function inlineComment()
362			{
363			do {
364	36		$this->next();
365	36		if ($this->ch === "\n" \|\| $this->ch === "\r") {
366	33		$this->next();
367
368	33		return;
369			}
370	36		} while ($this->ch !== null);
371	3		}
372
373			/**
374			* Skip a block comment, assuming this is one.
375			*
376			* The current character should be the * character in the /* pair that begins this block comment.
377			* To finish the block comment, we look for an ending */ pair of characters,
378			* but we also watch for the end of text before the comment is terminated.
379			*/
380	21		private function blockComment()
381			{
382			do {
383	21		$this->next();
384	21		while ($this->ch === '*') {
385	18		$this->nextOrFail('*');
386	18		if ($this->ch === '/') {
387	18		$this->nextOrFail('/');
388
389	18		return;
390			}
391	2		}
392	21		} while ($this->ch !== null);
393
394	3		$this->throwSyntaxError('Unterminated block comment');
395			}
396
397			/**
398			* Skip a comment, whether inline or block-level, assuming this is one.
399			*/
400	57		private function comment()
401			{
402			// Comments always begin with a / character.
403	57		$this->nextOrFail('/');
404
405	57		if ($this->ch === '/') {
406	36		$this->inlineComment();
407	46		} elseif ($this->ch === '*') {
408	21		$this->blockComment();
409	12		} else {
410	3		$this->throwSyntaxError('Unrecognized comment');
411			}
412	51		}
413
414			/**
415			* Skip whitespace and comments.
416			*
417			* Note that we're detecting comments by only a single / character.
418			* This works since regular expressions are not valid JSON(5), but this will
419			* break if there are other valid values that begin with a / character!
420			*/
421	223		private function white()
422			{
423	223		while ($this->ch !== null) {
424	222		if ($this->ch === '/') {
425	57		$this->comment();
426	217		} elseif (preg_match('/[ \t\r\n\v\f\xA0\x{FEFF}]/u', $this->ch) === 1) {
427	164		$this->next();
428	94		} else {
429	213		return;
430			}
431	98		}
432	141		}
433
434			/**
435			* Matches true, false, null, etc
436			*/
437	67		private function word()
438			{
439	67		switch ($this->ch) {
440	67		case 't':
441	21		$this->nextOrFail('t');
442	21		$this->nextOrFail('r');
443	21		$this->nextOrFail('u');
444	21		$this->nextOrFail('e');
445	21		return true;
446	52		case 'f':
447	9		$this->nextOrFail('f');
448	9		$this->nextOrFail('a');
449	9		$this->nextOrFail('l');
450	9		$this->nextOrFail('s');
451	9		$this->nextOrFail('e');
452	9		return false;
453	43		case 'n':
454	15		$this->nextOrFail('n');
455	15		$this->nextOrFail('u');
456	15		$this->nextOrFail('l');
457	15		$this->nextOrFail('l');
458	15		return null;
459	28		case 'I':
460	15		$this->nextOrFail('I');
461	15		$this->nextOrFail('n');
462	12		$this->nextOrFail('f');
463	12		$this->nextOrFail('i');
464	12		$this->nextOrFail('n');
465	12		$this->nextOrFail('i');
466	12		$this->nextOrFail('t');
467	12		$this->nextOrFail('y');
468	12		return INF;
469	13		case 'N':
470	6		$this->nextOrFail('N');
471	6		$this->nextOrFail('a');
472	3		$this->nextOrFail('N');
473	3		return NAN;
474	4		}
475
476	7		$this->throwSyntaxError('Unexpected ' . self::renderChar($this->ch));
477			}
478
479	27		private function arr()
480			{
481	27		$arr = [];
482
483	27		if ($this->ch === '[') {
484	27		if (++$this->depth > $this->maxDepth) {
485	3		$this->throwSyntaxError('Maximum stack depth exceeded');
486			}
487
488	27		$this->nextOrFail('[');
489	27		$this->white();
490	27		while ($this->ch !== null) {
491	27		if ($this->ch === ']') {
492	9		$this->nextOrFail(']');
493	9		$this->depth--;
494	9		return $arr; // Potentially empty array
495			}
496			// ES5 allows omitting elements in arrays, e.g. [,] and
497			// [,null]. We don't allow this in JSON5.
498	27		if ($this->ch === ',') {
499	6		$this->throwSyntaxError('Missing array element');
500			}
501
502	21		$arr[] = $this->value();
503
504	18		$this->white();
505			// If there's no comma after this value, this needs to
506			// be the end of the array.
507	18		if ($this->ch !== ',') {
508	9		$this->nextOrFail(']');
509	6		$this->depth--;
510	6		return $arr;
511			}
512	9		$this->nextOrFail(',');
513	9		$this->white();
514	6		}
515			}
516
517			$this->throwSyntaxError('Bad array');
518			}
519
520			/**
521			* Parse an object value
522			*/
523	72		private function obj()
524			{
525	72		$object = $this->associative ? [] : new \stdClass;
526
527	72		if ($this->ch === '{') {
528	72		if (++$this->depth > $this->maxDepth) {
529	3		$this->throwSyntaxError('Maximum stack depth exceeded');
530			}
531
532	72		$this->nextOrFail('{');
533	72		$this->white();
534	72		while ($this->ch) {
535	72		if ($this->ch === '}') {
536	18		$this->nextOrFail('}');
537	18		$this->depth--;
538	18		return $object; // Potentially empty object
539			}
540
541			// Keys can be unquoted. If they are, they need to be
542			// valid JS identifiers.
543	63		if ($this->ch === '"' \|\| $this->ch === "'") {
544	24		$key = $this->string();
545	16		} else {
546	42		$key = $this->identifier();
547			}
548
549	54		$this->white();
550	54		$this->nextOrFail(':');
551	51		if ($this->associative) {
552	39		$object[$key] = $this->value();
553	26		} else {
554	51		$object->{$key} = $this->value();
555			}
556	48		$this->white();
557			// If there's no comma after this pair, this needs to be
558			// the end of the object.
559	48		if ($this->ch !== ',') {
560	39		$this->nextOrFail('}');
561	33		$this->depth--;
562	33		return $object;
563			}
564	15		$this->nextOrFail(',');
565	15		$this->white();
566	10		}
567			}
568
569			$this->throwSyntaxError('Bad object');
570			}
571
572			/**
573			* Parse a JSON value.
574			*
575			* It could be an object, an array, a string, a number,
576			* or a word.
577			*/
578	223		private function value()
579			{
580	223		$this->white();
581	220		switch ($this->ch) {
582	220		case '{':
583	72		return $this->obj();
584	196		case '[':
585	27		return $this->arr();
586	187		case '"':
587	177		case "'":
588	57		return $this->string();
589	136		case '-':
590	126		case '+':
591	111		case '.':
592	45		return $this->number();
593	54		default:
594	94		return is_numeric($this->ch) ? $this->number() : $this->word();
595	54		}
596			}
597
598	92		private function throwSyntaxError($message)
599			{
600	92		throw new SyntaxError($message, $this->lineNumber, $this->columnNumber);
601			}
602
603	25		private static function renderChar($chr)
604			{
605	25		return $chr === null ? 'EOF' : "'" . $chr . "'";
606			}
607
608			/**
609			* @param string $ch
610			*
611			* @return string\|null
612			*/
613	18		private static function getEscapee($ch)
614			{
615			switch ($ch) {
616			// @codingStandardsIgnoreStart
617	18		case "'": return "'";
618	15		case '"': return '"';
619	15		case '\\': return '\\';
620	15		case '/': return '/';
621	15		case "\n": return '';
622	6		case 'b': return chr(8);
623	6		case 'f': return "\f";
624	6		case 'n': return "\n";
625	6		case 'r': return "\r";
626	6		case 't': return "\t";
627	3		default: return null;
628			// @codingStandardsIgnoreEnd
629	2		}
630			}
631
632			/**
633			* Returns everything from $this->at onwards.
634			*
635			* Utilizes a cache so we don't have to continuously parse through UTF-8
636			* data that was earlier in the string which we don't even care about.
637			*
638			* @return string
639			*/
640	103		private function getRemainder()
641			{
642	103		if ($this->remainderCacheAt === $this->at) {
643	25		return $this->remainderCache;
644			}
645
646	97		$subject = mb_substr($this->remainderCache, $this->at - $this->remainderCacheAt);
647	97		$this->remainderCache = $subject;
648	97		$this->remainderCacheAt = $this->at;
649
650	97		return $subject;
651			}
652			}
653

colinodell / json5

Pull Request — master (#2)

Json5Decoder::next() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like