StringUtils::isUtf8() - Code Metrics - Inspection of "Daily Inspection: Update OOjs UI to v0.16.5" - wikimedia/mediawiki - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Branch — master (d58858)

unknown

created 2016-04-07 17:09 UTC

StringUtils::isUtf8() A

↳ Parent: StringUtils

Complexity

Conditions	4
Paths	6

Size

Total Lines	13
Code Lines	7

Duplication

Lines	0
Ratio	0 %

Importance

Changes	1
Bugs	0	Features	0

Metric	Value
c	1
b	0
f	0
dl	0
loc	13
rs	9.2
cc	4
eloc	7
nc	6
nop	1

<?php
/**
 * Methods to play with strings.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License along
 * with this program; if not, write to the Free Software Foundation, Inc.,
 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
 * http://www.gnu.org/copyleft/gpl.html
 *
 * @file
 */

/**
 * A collection of static methods to play with strings.
 */
class StringUtils {
	/**
	 * Test whether a string is valid UTF-8.
	 *
	 * The function check for invalid byte sequences, overlong encoding but
	 * not for different normalisations.
	 *
	 * @note In MediaWiki 1.21, this function did not provide proper UTF-8 validation.
	 * In particular, the pure PHP code path did not in fact check for overlong forms.
	 * Beware of this when backporting code to that version of MediaWiki.
	 *
	 * @since 1.21
	 * @param string $value String to check
	 * @return bool Whether the given $value is a valid UTF-8 encoded string
	 */
	static function isUtf8( $value ) {
		$value = (string)$value;

		// Before PHP 5.4, values above U+10FFFF are incorrectly allowed, so we have to
		// check for them separately.
		static $newPHP;
		if ( $newPHP === null ) {
			$newPHP = !mb_check_encoding( "\xf4\x90\x80\x80", 'UTF-8' );
		}

		return mb_check_encoding( $value, 'UTF-8' ) &&
			( $newPHP || preg_match( "/\xf4[\x90-\xbf]|[\xf5-\xff]/S", $value ) === 0 );
	}

	/**
	 * Perform an operation equivalent to `preg_replace()`
	 *
	 * Matches this code:
	 *
	 *     preg_replace( "!$startDelim(.*?)$endDelim!", $replace, $subject );
	 *
	 * ..except that it's worst-case O(N) instead of O(N^2). Compared to delimiterReplace(), this
	 * implementation is fast but memory-hungry and inflexible. The memory requirements are such
	 * that I don't recommend using it on anything but guaranteed small chunks of text.
	 *
	 * @param string $startDelim
	 * @param string $endDelim
	 * @param string $replace
	 * @param string $subject
	 * @return string
	 */
	static function hungryDelimiterReplace( $startDelim, $endDelim, $replace, $subject ) {
		$segments = explode( $startDelim, $subject );
		$output = array_shift( $segments );
		foreach ( $segments as $s ) {
			$endDelimPos = strpos( $s, $endDelim );
			if ( $endDelimPos === false ) {
				$output .= $startDelim . $s;
			} else {
				$output .= $replace . substr( $s, $endDelimPos + strlen( $endDelim ) );
			}
		}

		return $output;
	}

	/**
	 * Perform an operation equivalent to `preg_replace_callback()`
	 *
	 * Matches this code:
	 *
	 *     preg_replace_callback( "!$startDelim(.*)$endDelim!s$flags", $callback, $subject );
	 *
	 * If the start delimiter ends with an initial substring of the end delimiter,
	 * e.g. in the case of C-style comments, the behavior differs from the model
	 * regex. In this implementation, the end must share no characters with the
	 * start, so e.g. `/*\/` is not considered to be both the start and end of a
	 * comment. `/*\/xy/*\/` is considered to be a single comment with contents `/xy/`.
	 *
	 * The implementation of delimiterReplaceCallback() is slower than hungryDelimiterReplace()
	 * but uses far less memory. The delimiters are literal strings, not regular expressions.
	 *
	 * @param string $startDelim Start delimiter
	 * @param string $endDelim End delimiter
	 * @param callable $callback Function to call on each match
	 * @param string $subject
	 * @param string $flags Regular expression flags
	 * @throws InvalidArgumentException
	 * @return string
	 */
	static function delimiterReplaceCallback( $startDelim, $endDelim, $callback,
		$subject, $flags = ''
	) {
		$inputPos = 0;
		$outputPos = 0;
		$output = '';
		$foundStart = false;
		$encStart = preg_quote( $startDelim, '!' );
		$encEnd = preg_quote( $endDelim, '!' );
		$strcmp = strpos( $flags, 'i' ) === false ? 'strcmp' : 'strcasecmp';
		$endLength = strlen( $endDelim );
		$m = [];

		while ( $inputPos < strlen( $subject ) &&
			preg_match( "!($encStart)|($encEnd)!S$flags", $subject, $m, PREG_OFFSET_CAPTURE, $inputPos )
		) {
			$tokenOffset = $m[0][1];
			if ( $m[1][0] != '' ) {
				if ( $foundStart &&
					$strcmp( $endDelim, substr( $subject, $tokenOffset, $endLength ) ) == 0
				) {
					# An end match is present at the same location
					$tokenType = 'end';
					$tokenLength = $endLength;
				} else {
					$tokenType = 'start';
					$tokenLength = strlen( $m[0][0] );
				}
			} elseif ( $m[2][0] != '' ) {
				$tokenType = 'end';
				$tokenLength = strlen( $m[0][0] );
			} else {
				throw new InvalidArgumentException( 'Invalid delimiter given to ' . __METHOD__ );
			}

			if ( $tokenType == 'start' ) {
				# Only move the start position if we haven't already found a start
				# This means that START START END matches outer pair
				if ( !$foundStart ) {
					# Found start
					$inputPos = $tokenOffset + $tokenLength;
					# Write out the non-matching section
					$output .= substr( $subject, $outputPos, $tokenOffset - $outputPos );
					$outputPos = $tokenOffset;
					$contentPos = $inputPos;
					$foundStart = true;
				} else {
					# Move the input position past the *first character* of START,
					# to protect against missing END when it overlaps with START
					$inputPos = $tokenOffset + 1;
				}
			} elseif ( $tokenType == 'end' ) {
				if ( $foundStart ) {
					# Found match
					$output .= call_user_func( $callback, [
						substr( $subject, $outputPos, $tokenOffset + $tokenLength - $outputPos ),
						substr( $subject, $contentPos, $tokenOffset - $contentPos )
function myFunction($a) {
    switch ($a) {
        case 'foo':
            $x = 1;
            break;

        case 'bar':
            $x = 2;
            break;
    }

    // $x is potentially undefined here.
    echo $x;
}
					] );
					$foundStart = false;
				} else {
					# Non-matching end, write it out
					$output .= substr( $subject, $inputPos, $tokenOffset + $tokenLength - $outputPos );
				}
				$inputPos = $outputPos = $tokenOffset + $tokenLength;
			} else {
				throw new InvalidArgumentException( 'Invalid delimiter given to ' . __METHOD__ );
			}
		}
		if ( $outputPos < strlen( $subject ) ) {
			$output .= substr( $subject, $outputPos );
		}

		return $output;
	}

	/**
	 * Perform an operation equivalent to `preg_replace()` with flags.
	 *
	 * Matches this code:
	 *
	 *     preg_replace( "!$startDelim(.*)$endDelim!$flags", $replace, $subject );
	 *
	 * @param string $startDelim Start delimiter regular expression
	 * @param string $endDelim End delimiter regular expression
	 * @param string $replace Replacement string. May contain $1, which will be
	 *  replaced by the text between the delimiters
	 * @param string $subject String to search
	 * @param string $flags Regular expression flags
	 * @return string The string with the matches replaced
	 */
	static function delimiterReplace( $startDelim, $endDelim, $replace, $subject, $flags = '' ) {
		$replacer = new RegexlikeReplacer( $replace );

		return self::delimiterReplaceCallback( $startDelim, $endDelim,
			$replacer->cb(), $subject, $flags );
	}

	/**
	 * More or less "markup-safe" explode()
	 * Ignores any instances of the separator inside `<...>`
	 * @param string $separator
	 * @param string $text
	 * @return array
	 */
	static function explodeMarkup( $separator, $text ) {
		$placeholder = "\x00";

		// Remove placeholder instances
		$text = str_replace( $placeholder, '', $text );

		// Replace instances of the separator inside HTML-like tags with the placeholder
		$replacer = new DoubleReplacer( $separator, $placeholder );
		$cleaned = StringUtils::delimiterReplaceCallback( '<', '>', $replacer->cb(), $text );

		// Explode, then put the replaced separators back in
		$items = explode( $separator, $cleaned );
		foreach ( $items as $i => $str ) {
			$items[$i] = str_replace( $placeholder, $separator, $str );
		}

		return $items;
	}

	/**
	 * More or less "markup-safe" str_replace()
	 * Ignores any instances of the separator inside `<...>`
	 * @param string $search
	 * @param string $replace
	 * @param string $text
	 * @return string
	 */
	static function replaceMarkup( $search, $replace, $text ) {
		$placeholder = "\x00";

		// Remove placeholder instances
		$text = str_replace( $placeholder, '', $text );

		// Replace instances of the separator inside HTML-like tags with the placeholder
		$replacer = new DoubleReplacer( $search, $placeholder );
		$cleaned = StringUtils::delimiterReplaceCallback( '<', '>', $replacer->cb(), $text );

		// Explode, then put the replaced separators back in
		$cleaned = str_replace( $search, $replace, $cleaned );
		$text = str_replace( $placeholder, $search, $cleaned );

		return $text;
	}

	/**
	 * Escape a string to make it suitable for inclusion in a preg_replace()
	 * replacement parameter.
	 *
	 * @param string $string
	 * @return string
	 */
	static function escapeRegexReplacement( $string ) {
		$string = str_replace( '\\', '\\\\', $string );
		$string = str_replace( '$', '\\$', $string );
		return $string;
	}

	/**
	 * Workalike for explode() with limited memory usage.
	 *
	 * @param string $separator
	 * @param string $subject
	 * @return ArrayIterator|ExplodeIterator
	 */
	static function explode( $separator, $subject ) {
		if ( substr_count( $subject, $separator ) > 1000 ) {
			return new ExplodeIterator( $separator, $subject );
		} else {
			return new ArrayIterator( explode( $separator, $subject ) );
		}
	}
}


1			<?php
2			/**
3			* Methods to play with strings.
4			*
5			* This program is free software; you can redistribute it and/or modify
6			* it under the terms of the GNU General Public License as published by
7			* the Free Software Foundation; either version 2 of the License, or
8			* (at your option) any later version.
9			*
10			* This program is distributed in the hope that it will be useful,
11			* but WITHOUT ANY WARRANTY; without even the implied warranty of
12			* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13			* GNU General Public License for more details.
14			*
15			* You should have received a copy of the GNU General Public License along
16			* with this program; if not, write to the Free Software Foundation, Inc.,
17			* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18			* http://www.gnu.org/copyleft/gpl.html
19			*
20			* @file
21			*/
22
23			/**
24			* A collection of static methods to play with strings.
25			*/
26			class StringUtils {
27			/**
28			* Test whether a string is valid UTF-8.
29			*
30			* The function check for invalid byte sequences, overlong encoding but
31			* not for different normalisations.
32			*
33			* @note In MediaWiki 1.21, this function did not provide proper UTF-8 validation.
34			* In particular, the pure PHP code path did not in fact check for overlong forms.
35			* Beware of this when backporting code to that version of MediaWiki.
36			*
37			* @since 1.21
38			* @param string $value String to check
39			* @return bool Whether the given $value is a valid UTF-8 encoded string
40			*/
41			static function isUtf8( $value ) {
42			$value = (string)$value;
43
44			// Before PHP 5.4, values above U+10FFFF are incorrectly allowed, so we have to
45			// check for them separately.
46			static $newPHP;
47			if ( $newPHP === null ) {
48			$newPHP = !mb_check_encoding( "\xf4\x90\x80\x80", 'UTF-8' );
49			}
50
51			return mb_check_encoding( $value, 'UTF-8' ) &&
52			( $newPHP \|\| preg_match( "/\xf4[\x90-\xbf]\|[\xf5-\xff]/S", $value ) === 0 );
53			}
54
55			/**
56			* Perform an operation equivalent to `preg_replace()`
57			*
58			* Matches this code:
59			*
60			* preg_replace( "!$startDelim(.*?)$endDelim!", $replace, $subject );
61			*
62			* ..except that it's worst-case O(N) instead of O(N^2). Compared to delimiterReplace(), this
63			* implementation is fast but memory-hungry and inflexible. The memory requirements are such
64			* that I don't recommend using it on anything but guaranteed small chunks of text.
65			*
66			* @param string $startDelim
67			* @param string $endDelim
68			* @param string $replace
69			* @param string $subject
70			* @return string
71			*/
72			static function hungryDelimiterReplace( $startDelim, $endDelim, $replace, $subject ) {
73			$segments = explode( $startDelim, $subject );
74			$output = array_shift( $segments );
75			foreach ( $segments as $s ) {
76			$endDelimPos = strpos( $s, $endDelim );
77			if ( $endDelimPos === false ) {
78			$output .= $startDelim . $s;
79			} else {
80			$output .= $replace . substr( $s, $endDelimPos + strlen( $endDelim ) );
81			}
82			}
83
84			return $output;
85			}
86
87			/**
88			* Perform an operation equivalent to `preg_replace_callback()`
89			*
90			* Matches this code:
91			*
92			* preg_replace_callback( "!$startDelim(.*)$endDelim!s$flags", $callback, $subject );
93			*
94			* If the start delimiter ends with an initial substring of the end delimiter,
95			* e.g. in the case of C-style comments, the behavior differs from the model
96			* regex. In this implementation, the end must share no characters with the
97			* start, so e.g. `/*\/` is not considered to be both the start and end of a
98			* comment. `/\/xy/\/` is considered to be a single comment with contents `/xy/`.
99			*
100			* The implementation of delimiterReplaceCallback() is slower than hungryDelimiterReplace()
101			* but uses far less memory. The delimiters are literal strings, not regular expressions.
102			*
103			* @param string $startDelim Start delimiter
104			* @param string $endDelim End delimiter
105			* @param callable $callback Function to call on each match
106			* @param string $subject
107			* @param string $flags Regular expression flags
108			* @throws InvalidArgumentException
109			* @return string
110			*/
111			static function delimiterReplaceCallback( $startDelim, $endDelim, $callback,
112			$subject, $flags = ''
113			) {
114			$inputPos = 0;
115			$outputPos = 0;
116			$output = '';
117			$foundStart = false;
118			$encStart = preg_quote( $startDelim, '!' );
119			$encEnd = preg_quote( $endDelim, '!' );
120			$strcmp = strpos( $flags, 'i' ) === false ? 'strcmp' : 'strcasecmp';
121			$endLength = strlen( $endDelim );
122			$m = [];
123
124			while ( $inputPos < strlen( $subject ) &&
125			preg_match( "!($encStart)\|($encEnd)!S$flags", $subject, $m, PREG_OFFSET_CAPTURE, $inputPos )
126			) {
127			$tokenOffset = $m[0][1];
128			if ( $m[1][0] != '' ) {
129			if ( $foundStart &&
130			$strcmp( $endDelim, substr( $subject, $tokenOffset, $endLength ) ) == 0
131			) {
132			# An end match is present at the same location
133			$tokenType = 'end';
134			$tokenLength = $endLength;
135			} else {
136			$tokenType = 'start';
137			$tokenLength = strlen( $m[0][0] );
138			}
139			} elseif ( $m[2][0] != '' ) {
140			$tokenType = 'end';
141			$tokenLength = strlen( $m[0][0] );
142			} else {
143			throw new InvalidArgumentException( 'Invalid delimiter given to ' . __METHOD__ );
144			}
145
146			if ( $tokenType == 'start' ) {
147			# Only move the start position if we haven't already found a start
148			# This means that START START END matches outer pair
149			if ( !$foundStart ) {
150			# Found start
151			$inputPos = $tokenOffset + $tokenLength;
152			# Write out the non-matching section
153			$output .= substr( $subject, $outputPos, $tokenOffset - $outputPos );
154			$outputPos = $tokenOffset;
155			$contentPos = $inputPos;
156			$foundStart = true;
157			} else {
158			# Move the input position past the first character of START,
159			# to protect against missing END when it overlaps with START
160			$inputPos = $tokenOffset + 1;
161			}
162			} elseif ( $tokenType == 'end' ) {
163			if ( $foundStart ) {
164			# Found match
165			$output .= call_user_func( $callback, [
166			substr( $subject, $outputPos, $tokenOffset + $tokenLength - $outputPos ),
167			substr( $subject, $contentPos, $tokenOffset - $contentPos )
			0 ignored issues – show Bug introduced 2016-01-16 18:00 UTC by Report Bug Copy Issue Report The variable `$contentPos` does not seem to be defined for all execution paths leading up to this point. If you define a variable conditionally, it can happen that it is not defined for all execution paths. Let’s take a look at an example: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } // $x is potentially undefined here. echo $x; } In the above example, the variable `$x` is defined if you pass “foo” or “bar” as argument for `$a`. However, since the `switch` statement has no default case statement, if you pass any other value, the variable `$x` would be undefined. Available Fixes Check for existence of the variable explicitly: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } if (isset($x)) { // Make sure it's always set. echo $x; } } Define a default value for the variable: function myFunction($a) { $x = ''; // Set a default which gets overridden for certain paths. switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } echo $x; } Add a value for the missing path: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; // We add support for the missing case. default: $x = ''; break; } echo $x; } Loading history...
168			] );
169			$foundStart = false;
170			} else {
171			# Non-matching end, write it out
172			$output .= substr( $subject, $inputPos, $tokenOffset + $tokenLength - $outputPos );
173			}
174			$inputPos = $outputPos = $tokenOffset + $tokenLength;
175			} else {
176			throw new InvalidArgumentException( 'Invalid delimiter given to ' . __METHOD__ );
177			}
178			}
179			if ( $outputPos < strlen( $subject ) ) {
180			$output .= substr( $subject, $outputPos );
181			}
182
183			return $output;
184			}
185
186			/**
187			* Perform an operation equivalent to `preg_replace()` with flags.
188			*
189			* Matches this code:
190			*
191			* preg_replace( "!$startDelim(.*)$endDelim!$flags", $replace, $subject );
192			*
193			* @param string $startDelim Start delimiter regular expression
194			* @param string $endDelim End delimiter regular expression
195			* @param string $replace Replacement string. May contain $1, which will be
196			* replaced by the text between the delimiters
197			* @param string $subject String to search
198			* @param string $flags Regular expression flags
199			* @return string The string with the matches replaced
200			*/
201			static function delimiterReplace( $startDelim, $endDelim, $replace, $subject, $flags = '' ) {
202			$replacer = new RegexlikeReplacer( $replace );
203
204			return self::delimiterReplaceCallback( $startDelim, $endDelim,
205			$replacer->cb(), $subject, $flags );
206			}
207
208			/**
209			* More or less "markup-safe" explode()
210			* Ignores any instances of the separator inside `<...>`
211			* @param string $separator
212			* @param string $text
213			* @return array
214			*/
215			static function explodeMarkup( $separator, $text ) {
216			$placeholder = "\x00";
217
218			// Remove placeholder instances
219			$text = str_replace( $placeholder, '', $text );
220
221			// Replace instances of the separator inside HTML-like tags with the placeholder
222			$replacer = new DoubleReplacer( $separator, $placeholder );
223			$cleaned = StringUtils::delimiterReplaceCallback( '<', '>', $replacer->cb(), $text );
224
225			// Explode, then put the replaced separators back in
226			$items = explode( $separator, $cleaned );
227			foreach ( $items as $i => $str ) {
228			$items[$i] = str_replace( $placeholder, $separator, $str );
229			}
230
231			return $items;
232			}
233
234			/**
235			* More or less "markup-safe" str_replace()
236			* Ignores any instances of the separator inside `<...>`
237			* @param string $search
238			* @param string $replace
239			* @param string $text
240			* @return string
241			*/
242			static function replaceMarkup( $search, $replace, $text ) {
243			$placeholder = "\x00";
244
245			// Remove placeholder instances
246			$text = str_replace( $placeholder, '', $text );
247
248			// Replace instances of the separator inside HTML-like tags with the placeholder
249			$replacer = new DoubleReplacer( $search, $placeholder );
250			$cleaned = StringUtils::delimiterReplaceCallback( '<', '>', $replacer->cb(), $text );
251
252			// Explode, then put the replaced separators back in
253			$cleaned = str_replace( $search, $replace, $cleaned );
254			$text = str_replace( $placeholder, $search, $cleaned );
255
256			return $text;
257			}
258
259			/**
260			* Escape a string to make it suitable for inclusion in a preg_replace()
261			* replacement parameter.
262			*
263			* @param string $string
264			* @return string
265			*/
266			static function escapeRegexReplacement( $string ) {
267			$string = str_replace( '\\', '\\\\', $string );
268			$string = str_replace( '$', '\\$', $string );
269			return $string;
270			}
271
272			/**
273			* Workalike for explode() with limited memory usage.
274			*
275			* @param string $separator
276			* @param string $subject
277			* @return ArrayIterator\|ExplodeIterator
278			*/
279			static function explode( $separator, $subject ) {
280			if ( substr_count( $subject, $separator ) > 1000 ) {
281			return new ExplodeIterator( $separator, $subject );
282			} else {
283			return new ArrayIterator( explode( $separator, $subject ) );
284			}
285			}
286			}
287

wikimedia / mediawiki

Branch — master (d58858)

StringUtils::isUtf8() A

Complexity

Size

Duplication

Importance

Available Fixes

Duplication Side-by-Side

Filter issues like