ParallelRegex::apply() - Code Metrics - Inspection of "PHP 8: rename match() to apply() in Lexer #3298" - splitbrain/dokuwiki - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( bc45a2...fe2e97 )

by Andreas

created 2020-12-03 20:48 UTC

ParallelRegex::apply() B

↳ Parent: ParallelRegex

Complexity

Conditions	6
Paths	5

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	6
nc	5
nop	2
dl	0
loc	20
rs	8.9777
c	0
b	0
f	0

<?php
/**
 * Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/
 * For an intro to the Lexer see:
 * https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes
 *
 * @author Marcus Baker http://www.lastcraft.com
 */

namespace dokuwiki\Parsing\Lexer;

/**
 * Compounded regular expression.
 *
 * Any of the contained patterns could match and when one does it's label is returned.
 */
class ParallelRegex
{
    /** @var string[] patterns to match */
    protected $patterns;
    /** @var string[] labels for above patterns */
    protected $labels;
    /** @var string the compound regex matching all patterns */
    protected $regex;
    /** @var bool case sensitive matching? */
    protected $case;

    /**
     * Constructor. Starts with no patterns.
     *
     * @param boolean $case    True for case sensitive, false
     *                         for insensitive.
     */
    public function __construct($case)
    {
        $this->case = $case;
        $this->patterns = array();
        $this->labels = array();
        $this->regex = null;
    }

    /**
     * Adds a pattern with an optional label.
     *
     * @param mixed       $pattern Perl style regex. Must be UTF-8
     *                             encoded. If its a string, the (, )
     *                             lose their meaning unless they
     *                             form part of a lookahead or
     *                             lookbehind assertation.
     * @param bool|string $label   Label of regex to be returned
     *                             on a match. Label must be ASCII
     */
    public function addPattern($pattern, $label = true)
    {
        $count = count($this->patterns);
        $this->patterns[$count] = $pattern;
        $this->labels[$count] = $label;
        $this->regex = null;
    }

    /**
     * Attempts to match all patterns at once against a string.
     *
     * @param string $subject      String to match against.
     * @param string $match        First matched portion of
     *                             subject.
     * @return bool|string         False if no match found, label if label exists, true if not
     */
    public function apply($subject, &$match)
    {
        if (count($this->patterns) == 0) {
            return false;
        }
        if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
            $match = "";
            return false;
        }

        $match = $matches[0];
        $size = count($matches);
        // FIXME this could be made faster by storing the labels as keys in a hashmap
        for ($i = 1; $i < $size; $i++) {
            if ($matches[$i] && isset($this->labels[$i - 1])) {
                return $this->labels[$i - 1];
            }
        }
        return true;
    }

    /**
     * Attempts to split the string against all patterns at once
     *
     * @param string $subject      String to match against.
     * @param array $split         The split result: array containing, pre-match, match & post-match strings
     * @return boolean             True on success.
     *
     * @author Christopher Smith <[email protected]>
     */
    public function split($subject, &$split)
    {
        if (count($this->patterns) == 0) {
            return false;
        }

        if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
            if (function_exists('preg_last_error')) {
                $err = preg_last_error();
                switch ($err) {
                    case PREG_BACKTRACK_LIMIT_ERROR:
                        msg('A PCRE backtrack error occured. Try to increase the pcre.backtrack_limit in php.ini', -1);
                        break;
                    case PREG_RECURSION_LIMIT_ERROR:
                        msg('A PCRE recursion error occured. Try to increase the pcre.recursion_limit in php.ini', -1);
                        break;
                    case PREG_BAD_UTF8_ERROR:
                        msg('A PCRE UTF-8 error occured. This might be caused by a faulty plugin', -1);
                        break;
                    case PREG_INTERNAL_ERROR:
                        msg('A PCRE internal error occured. This might be caused by a faulty plugin', -1);
                        break;
                }
            }

            $split = array($subject, "", "");
            return false;
        }

        $idx = count($matches)-2;
        list($pre, $post) = preg_split($this->patterns[$idx].$this->getPerlMatchingFlags(), $subject, 2);
        $split = array($pre, $matches[0], $post);

        return isset($this->labels[$idx]) ? $this->labels[$idx] : true;
    }

    /**
     * Compounds the patterns into a single
     * regular expression separated with the
     * "or" operator. Caches the regex.
     * Will automatically escape (, ) and / tokens.
     *
     * @return null|string
     */
    protected function getCompoundedRegex()
    {
        if ($this->regex == null) {
            $cnt = count($this->patterns);
            for ($i = 0; $i < $cnt; $i++) {
                /*
                 * decompose the input pattern into "(", "(?", ")",
                 * "[...]", "[]..]", "[^]..]", "[...[:...:]..]", "\x"...
                 * elements.
                 */
                preg_match_all('/\\\\.|' .
                               '\(\?|' .
                               '[()]|' .
                               '\[\^?\]?(?:\\\\.|\[:[^]]*:\]|[^]\\\\])*\]|' .
                               '[^[()\\\\]+/', $this->patterns[$i], $elts);

                $pattern = "";
                $level = 0;

                foreach ($elts[0] as $elt) {
                    /*
                     * for "(", ")" remember the nesting level, add "\"
                     * only to the non-"(?" ones.
                     */

                    switch ($elt) {
                        case '(':
                            $pattern .= '\(';
                            break;
                        case ')':
                            if ($level > 0)
                                $level--; /* closing (? */
                            else $pattern .= '\\';
                            $pattern .= ')';
                            break;
                        case '(?':
                            $level++;
                            $pattern .= '(?';
                            break;
                        default:
                            if (substr($elt, 0, 1) == '\\')
                                $pattern .= $elt;
                            else $pattern .= str_replace('/', '\/', $elt);
                    }
                }
                $this->patterns[$i] = "($pattern)";
            }
            $this->regex = "/" . implode("|", $this->patterns) . "/" . $this->getPerlMatchingFlags();
        }
        return $this->regex;
    }

    /**
     * Accessor for perl regex mode flags to use.
     * @return string       Perl regex flags.
     */
    protected function getPerlMatchingFlags()
    {
        return ($this->case ? "msS" : "msSi");
    }
}


1			<?php
2			/**
3			* Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/
4			* For an intro to the Lexer see:
5			* https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes
6			*
7			* @author Marcus Baker http://www.lastcraft.com
8			*/
9
10			namespace dokuwiki\Parsing\Lexer;
11
12			/**
13			* Compounded regular expression.
14			*
15			* Any of the contained patterns could match and when one does it's label is returned.
16			*/
17			class ParallelRegex
18			{
19			/** @var string[] patterns to match */
20			protected $patterns;
21			/** @var string[] labels for above patterns */
22			protected $labels;
23			/** @var string the compound regex matching all patterns */
24			protected $regex;
25			/** @var bool case sensitive matching? */
26			protected $case;
27
28			/**
29			* Constructor. Starts with no patterns.
30			*
31			* @param boolean $case True for case sensitive, false
32			* for insensitive.
33			*/
34			public function __construct($case)
35			{
36			$this->case = $case;
37			$this->patterns = array();
38			$this->labels = array();
39			$this->regex = null;
40			}
41
42			/**
43			* Adds a pattern with an optional label.
44			*
45			* @param mixed $pattern Perl style regex. Must be UTF-8
46			* encoded. If its a string, the (, )
47			* lose their meaning unless they
48			* form part of a lookahead or
49			* lookbehind assertation.
50			* @param bool\|string $label Label of regex to be returned
51			* on a match. Label must be ASCII
52			*/
53			public function addPattern($pattern, $label = true)
54			{
55			$count = count($this->patterns);
56			$this->patterns[$count] = $pattern;
57			$this->labels[$count] = $label;
58			$this->regex = null;
59			}
60
61			/**
62			* Attempts to match all patterns at once against a string.
63			*
64			* @param string $subject String to match against.
65			* @param string $match First matched portion of
66			* subject.
67			* @return bool\|string False if no match found, label if label exists, true if not
68			*/
69			public function apply($subject, &$match)
70			{
71			if (count($this->patterns) == 0) {
72			return false;
73			}
74			if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
75			$match = "";
76			return false;
77			}
78
79			$match = $matches[0];
80			$size = count($matches);
81			// FIXME this could be made faster by storing the labels as keys in a hashmap
82			for ($i = 1; $i < $size; $i++) {
83			if ($matches[$i] && isset($this->labels[$i - 1])) {
84			return $this->labels[$i - 1];
85			}
86			}
87			return true;
88			}
89
90			/**
91			* Attempts to split the string against all patterns at once
92			*
93			* @param string $subject String to match against.
94			* @param array $split The split result: array containing, pre-match, match & post-match strings
95			* @return boolean True on success.
96			*
97			* @author Christopher Smith <[email protected]>
98			*/
99			public function split($subject, &$split)
100			{
101			if (count($this->patterns) == 0) {
102			return false;
103			}
104
105			if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
106			if (function_exists('preg_last_error')) {
107			$err = preg_last_error();
108			switch ($err) {
109			case PREG_BACKTRACK_LIMIT_ERROR:
110			msg('A PCRE backtrack error occured. Try to increase the pcre.backtrack_limit in php.ini', -1);
111			break;
112			case PREG_RECURSION_LIMIT_ERROR:
113			msg('A PCRE recursion error occured. Try to increase the pcre.recursion_limit in php.ini', -1);
114			break;
115			case PREG_BAD_UTF8_ERROR:
116			msg('A PCRE UTF-8 error occured. This might be caused by a faulty plugin', -1);
117			break;
118			case PREG_INTERNAL_ERROR:
119			msg('A PCRE internal error occured. This might be caused by a faulty plugin', -1);
120			break;
121			}
122			}
123
124			$split = array($subject, "", "");
125			return false;
126			}
127
128			$idx = count($matches)-2;
129			list($pre, $post) = preg_split($this->patterns[$idx].$this->getPerlMatchingFlags(), $subject, 2);
130			$split = array($pre, $matches[0], $post);
131
132			return isset($this->labels[$idx]) ? $this->labels[$idx] : true;
133			}
134
135			/**
136			* Compounds the patterns into a single
137			* regular expression separated with the
138			* "or" operator. Caches the regex.
139			* Will automatically escape (, ) and / tokens.
140			*
141			* @return null\|string
142			*/
143			protected function getCompoundedRegex()
144			{
145			if ($this->regex == null) {
146			$cnt = count($this->patterns);
147			for ($i = 0; $i < $cnt; $i++) {
148			/*
149			* decompose the input pattern into "(", "(?", ")",
150			* "[...]", "[]..]", "[^]..]", "[...[:...:]..]", "\x"...
151			* elements.
152			*/
153			preg_match_all('/\\\\.\|' .
154			'\(\?\|' .
155			'[()]\|' .
156			'\[\^?\]?(?:\\\\.\|\[:[^]]:\]\|[^]\\\\])\]\|' .
157			'[^[()\\\\]+/', $this->patterns[$i], $elts);
158
159			$pattern = "";
160			$level = 0;
161
162			foreach ($elts[0] as $elt) {
163			/*
164			* for "(", ")" remember the nesting level, add "\"
165			* only to the non-"(?" ones.
166			*/
167
168			switch ($elt) {
169			case '(':
170			$pattern .= '\(';
171			break;
172			case ')':
173			if ($level > 0)
174			$level--; /* closing (? */
175			else $pattern .= '\\';
176			$pattern .= ')';
177			break;
178			case '(?':
179			$level++;
180			$pattern .= '(?';
181			break;
182			default:
183			if (substr($elt, 0, 1) == '\\')
184			$pattern .= $elt;
185			else $pattern .= str_replace('/', '\/', $elt);
186			}
187			}
188			$this->patterns[$i] = "($pattern)";
189			}
190			$this->regex = "/" . implode("\|", $this->patterns) . "/" . $this->getPerlMatchingFlags();
191			}
192			return $this->regex;
193			}
194
195			/**
196			* Accessor for perl regex mode flags to use.
197			* @return string Perl regex flags.
198			*/
199			protected function getPerlMatchingFlags()
200			{
201			return ($this->case ? "msS" : "msSi");
202			}
203			}
204

splitbrain / dokuwiki

Push — master ( bc45a2...fe2e97 )

ParallelRegex::apply() B

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like