PageRenderTime 130ms CodeModel.GetById 5ms app.highlight 100ms RepoModel.GetById 1ms app.codeStats 1ms

/wp-includes/formatting.php

https://github.com/yanickouellet/WordPress
PHP | 3393 lines | 2143 code | 268 blank | 982 comment | 277 complexity | 591d4f3358331c593515e14508b5dcd9 MD5 | raw file
   1<?php
   2/**
   3 * Main WordPress Formatting API.
   4 *
   5 * Handles many functions for formatting output.
   6 *
   7 * @package WordPress
   8 */
   9
  10/**
  11 * Replaces common plain text characters into formatted entities
  12 *
  13 * As an example,
  14 * <code>
  15 * 'cause today's effort makes it worth tomorrow's "holiday"...
  16 * </code>
  17 * Becomes:
  18 * <code>
  19 * &#8217;cause today&#8217;s effort makes it worth tomorrow&#8217;s &#8220;holiday&#8221;&#8230;
  20 * </code>
  21 * Code within certain html blocks are skipped.
  22 *
  23 * @since 0.71
  24 * @uses $wp_cockneyreplace Array of formatted entities for certain common phrases
  25 *
  26 * @param string $text The text to be formatted
  27 * @return string The string replaced with html entities
  28 */
  29function wptexturize($text) {
  30	global $wp_cockneyreplace;
  31	static $static_characters, $static_replacements, $dynamic_characters, $dynamic_replacements,
  32		$default_no_texturize_tags, $default_no_texturize_shortcodes;
  33
  34	// No need to set up these static variables more than once
  35	if ( ! isset( $static_characters ) ) {
  36		/* translators: opening curly double quote */
  37		$opening_quote = _x( '&#8220;', 'opening curly double quote' );
  38		/* translators: closing curly double quote */
  39		$closing_quote = _x( '&#8221;', 'closing curly double quote' );
  40
  41		/* translators: apostrophe, for example in 'cause or can't */
  42		$apos = _x( '&#8217;', 'apostrophe' );
  43
  44		/* translators: prime, for example in 9' (nine feet) */
  45		$prime = _x( '&#8242;', 'prime' );
  46		/* translators: double prime, for example in 9" (nine inches) */
  47		$double_prime = _x( '&#8243;', 'double prime' );
  48
  49		/* translators: opening curly single quote */
  50		$opening_single_quote = _x( '&#8216;', 'opening curly single quote' );
  51		/* translators: closing curly single quote */
  52		$closing_single_quote = _x( '&#8217;', 'closing curly single quote' );
  53
  54		/* translators: en dash */
  55		$en_dash = _x( '&#8211;', 'en dash' );
  56		/* translators: em dash */
  57		$em_dash = _x( '&#8212;', 'em dash' );
  58
  59		$default_no_texturize_tags = array('pre', 'code', 'kbd', 'style', 'script', 'tt');
  60		$default_no_texturize_shortcodes = array('code');
  61
  62		// if a plugin has provided an autocorrect array, use it
  63		if ( isset($wp_cockneyreplace) ) {
  64			$cockney = array_keys($wp_cockneyreplace);
  65			$cockneyreplace = array_values($wp_cockneyreplace);
  66		} elseif ( "'" != $apos ) { // Only bother if we're doing a replacement.
  67			$cockney = array( "'tain't", "'twere", "'twas", "'tis", "'twill", "'til", "'bout", "'nuff", "'round", "'cause" );
  68			$cockneyreplace = array( $apos . "tain" . $apos . "t", $apos . "twere", $apos . "twas", $apos . "tis", $apos . "twill", $apos . "til", $apos . "bout", $apos . "nuff", $apos . "round", $apos . "cause" );
  69		} else {
  70			$cockney = $cockneyreplace = array();
  71		}
  72
  73		$static_characters = array_merge( array( '---', ' -- ', '--', ' - ', 'xn&#8211;', '...', '``', '\'\'', ' (tm)' ), $cockney );
  74		$static_replacements = array_merge( array( $em_dash, ' ' . $em_dash . ' ', $en_dash, ' ' . $en_dash . ' ', 'xn--', '&#8230;', $opening_quote, $closing_quote, ' &#8482;' ), $cockneyreplace );
  75
  76		$dynamic = array();
  77		if ( "'" != $apos ) {
  78			$dynamic[ '/\'(\d\d(?:&#8217;|\')?s)/' ] = $apos . '$1'; // '99's
  79			$dynamic[ '/\'(\d)/'                   ] = $apos . '$1'; // '99
  80		}
  81		if ( "'" != $opening_single_quote )
  82			$dynamic[ '/(\s|\A|[([{<]|")\'/'       ] = '$1' . $opening_single_quote; // opening single quote, even after (, {, <, [
  83		if ( '"' != $double_prime )
  84			$dynamic[ '/(\d)"/'                    ] = '$1' . $double_prime; // 9" (double prime)
  85		if ( "'" != $prime )
  86			$dynamic[ '/(\d)\'/'                   ] = '$1' . $prime; // 9' (prime)
  87		if ( "'" != $apos )
  88			$dynamic[ '/(\S)\'([^\'\s])/'          ] = '$1' . $apos . '$2'; // apostrophe in a word
  89		if ( '"' != $opening_quote )
  90			$dynamic[ '/(\s|\A|[([{<])"(?!\s)/'    ] = '$1' . $opening_quote . '$2'; // opening double quote, even after (, {, <, [
  91		if ( '"' != $closing_quote )
  92			$dynamic[ '/"(\s|\S|\Z)/'              ] = $closing_quote . '$1'; // closing double quote
  93		if ( "'" != $closing_single_quote )
  94			$dynamic[ '/\'([\s.]|\Z)/'             ] = $closing_single_quote . '$1'; // closing single quote
  95
  96		$dynamic[ '/\b(\d+)x(\d+)\b/'              ] = '$1&#215;$2'; // 9x9 (times)
  97
  98		$dynamic_characters = array_keys( $dynamic );
  99		$dynamic_replacements = array_values( $dynamic );
 100	}
 101
 102	// Transform into regexp sub-expression used in _wptexturize_pushpop_element
 103	// Must do this every time in case plugins use these filters in a context sensitive manner
 104	$no_texturize_tags = '(' . implode('|', apply_filters('no_texturize_tags', $default_no_texturize_tags) ) . ')';
 105	$no_texturize_shortcodes = '(' . implode('|', apply_filters('no_texturize_shortcodes', $default_no_texturize_shortcodes) ) . ')';
 106
 107	$no_texturize_tags_stack = array();
 108	$no_texturize_shortcodes_stack = array();
 109
 110	$textarr = preg_split('/(<.*>|\[.*\])/Us', $text, -1, PREG_SPLIT_DELIM_CAPTURE);
 111
 112	foreach ( $textarr as &$curl ) {
 113		if ( empty( $curl ) )
 114			continue;
 115
 116		// Only call _wptexturize_pushpop_element if first char is correct tag opening
 117		$first = $curl[0];
 118		if ( '<' === $first ) {
 119			_wptexturize_pushpop_element($curl, $no_texturize_tags_stack, $no_texturize_tags, '<', '>');
 120		} elseif ( '[' === $first ) {
 121			_wptexturize_pushpop_element($curl, $no_texturize_shortcodes_stack, $no_texturize_shortcodes, '[', ']');
 122		} elseif ( empty($no_texturize_shortcodes_stack) && empty($no_texturize_tags_stack) ) {
 123			// This is not a tag, nor is the texturization disabled static strings
 124			$curl = str_replace($static_characters, $static_replacements, $curl);
 125			// regular expressions
 126			$curl = preg_replace($dynamic_characters, $dynamic_replacements, $curl);
 127		}
 128		$curl = preg_replace('/&([^#])(?![a-zA-Z1-4]{1,8};)/', '&#038;$1', $curl);
 129	}
 130	return implode( '', $textarr );
 131}
 132
 133/**
 134 * Search for disabled element tags. Push element to stack on tag open and pop
 135 * on tag close. Assumes first character of $text is tag opening.
 136 *
 137 * @since 2.9.0
 138 * @access private
 139 *
 140 * @param string $text Text to check. First character is assumed to be $opening
 141 * @param array $stack Array used as stack of opened tag elements
 142 * @param string $disabled_elements Tags to match against formatted as regexp sub-expression
 143 * @param string $opening Tag opening character, assumed to be 1 character long
 144 * @param string $closing Tag closing character
 145 */
 146function _wptexturize_pushpop_element($text, &$stack, $disabled_elements, $opening = '<', $closing = '>') {
 147	// Check if it is a closing tag -- otherwise assume opening tag
 148	if (strncmp($opening . '/', $text, 2)) {
 149		// Opening? Check $text+1 against disabled elements
 150		if (preg_match('/^' . $disabled_elements . '\b/', substr($text, 1), $matches)) {
 151			/*
 152			 * This disables texturize until we find a closing tag of our type
 153			 * (e.g. <pre>) even if there was invalid nesting before that
 154			 *
 155			 * Example: in the case <pre>sadsadasd</code>"baba"</pre>
 156			 *          "baba" won't be texturize
 157			 */
 158
 159			array_push($stack, $matches[1]);
 160		}
 161	} else {
 162		// Closing? Check $text+2 against disabled elements
 163		$c = preg_quote($closing, '/');
 164		if (preg_match('/^' . $disabled_elements . $c . '/', substr($text, 2), $matches)) {
 165			$last = array_pop($stack);
 166
 167			// Make sure it matches the opening tag
 168			if ($last != $matches[1])
 169				array_push($stack, $last);
 170		}
 171	}
 172}
 173
 174/**
 175 * Replaces double line-breaks with paragraph elements.
 176 *
 177 * A group of regex replaces used to identify text formatted with newlines and
 178 * replace double line-breaks with HTML paragraph tags. The remaining
 179 * line-breaks after conversion become <<br />> tags, unless $br is set to '0'
 180 * or 'false'.
 181 *
 182 * @since 0.71
 183 *
 184 * @param string $pee The text which has to be formatted.
 185 * @param bool $br Optional. If set, this will convert all remaining line-breaks after paragraphing. Default true.
 186 * @return string Text which has been converted into correct paragraph tags.
 187 */
 188function wpautop($pee, $br = true) {
 189	$pre_tags = array();
 190
 191	if ( trim($pee) === '' )
 192		return '';
 193
 194	$pee = $pee . "\n"; // just to make things a little easier, pad the end
 195
 196	if ( strpos($pee, '<pre') !== false ) {
 197		$pee_parts = explode( '</pre>', $pee );
 198		$last_pee = array_pop($pee_parts);
 199		$pee = '';
 200		$i = 0;
 201
 202		foreach ( $pee_parts as $pee_part ) {
 203			$start = strpos($pee_part, '<pre');
 204
 205			// Malformed html?
 206			if ( $start === false ) {
 207				$pee .= $pee_part;
 208				continue;
 209			}
 210
 211			$name = "<pre wp-pre-tag-$i></pre>";
 212			$pre_tags[$name] = substr( $pee_part, $start ) . '</pre>';
 213
 214			$pee .= substr( $pee_part, 0, $start ) . $name;
 215			$i++;
 216		}
 217
 218		$pee .= $last_pee;
 219	}
 220
 221	$pee = preg_replace('|<br />\s*<br />|', "\n\n", $pee);
 222	// Space things out a little
 223	$allblocks = '(?:table|thead|tfoot|caption|col|colgroup|tbody|tr|td|th|div|dl|dd|dt|ul|ol|li|pre|select|option|form|map|area|blockquote|address|math|style|p|h[1-6]|hr|fieldset|noscript|legend|section|article|aside|hgroup|header|footer|nav|figure|figcaption|details|menu|summary)';
 224	$pee = preg_replace('!(<' . $allblocks . '[^>]*>)!', "\n$1", $pee);
 225	$pee = preg_replace('!(</' . $allblocks . '>)!', "$1\n\n", $pee);
 226	$pee = str_replace(array("\r\n", "\r"), "\n", $pee); // cross-platform newlines
 227	if ( strpos($pee, '<object') !== false ) {
 228		$pee = preg_replace('|\s*<param([^>]*)>\s*|', "<param$1>", $pee); // no pee inside object/embed
 229		$pee = preg_replace('|\s*</embed>\s*|', '</embed>', $pee);
 230	}
 231	$pee = preg_replace("/\n\n+/", "\n\n", $pee); // take care of duplicates
 232	// make paragraphs, including one at the end
 233	$pees = preg_split('/\n\s*\n/', $pee, -1, PREG_SPLIT_NO_EMPTY);
 234	$pee = '';
 235	foreach ( $pees as $tinkle )
 236		$pee .= '<p>' . trim($tinkle, "\n") . "</p>\n";
 237	$pee = preg_replace('|<p>\s*</p>|', '', $pee); // under certain strange conditions it could create a P of entirely whitespace
 238	$pee = preg_replace('!<p>([^<]+)</(div|address|form)>!', "<p>$1</p></$2>", $pee);
 239	$pee = preg_replace('!<p>\s*(</?' . $allblocks . '[^>]*>)\s*</p>!', "$1", $pee); // don't pee all over a tag
 240	$pee = preg_replace("|<p>(<li.+?)</p>|", "$1", $pee); // problem with nested lists
 241	$pee = preg_replace('|<p><blockquote([^>]*)>|i', "<blockquote$1><p>", $pee);
 242	$pee = str_replace('</blockquote></p>', '</p></blockquote>', $pee);
 243	$pee = preg_replace('!<p>\s*(</?' . $allblocks . '[^>]*>)!', "$1", $pee);
 244	$pee = preg_replace('!(</?' . $allblocks . '[^>]*>)\s*</p>!', "$1", $pee);
 245	if ( $br ) {
 246		$pee = preg_replace_callback('/<(script|style).*?<\/\\1>/s', '_autop_newline_preservation_helper', $pee);
 247		$pee = preg_replace('|(?<!<br />)\s*\n|', "<br />\n", $pee); // optionally make line breaks
 248		$pee = str_replace('<WPPreserveNewline />', "\n", $pee);
 249	}
 250	$pee = preg_replace('!(</?' . $allblocks . '[^>]*>)\s*<br />!', "$1", $pee);
 251	$pee = preg_replace('!<br />(\s*</?(?:p|li|div|dl|dd|dt|th|pre|td|ul|ol)[^>]*>)!', '$1', $pee);
 252	$pee = preg_replace( "|\n</p>$|", '</p>', $pee );
 253
 254	if ( !empty($pre_tags) )
 255		$pee = str_replace(array_keys($pre_tags), array_values($pre_tags), $pee);
 256
 257	return $pee;
 258}
 259
 260/**
 261 * Newline preservation help function for wpautop
 262 *
 263 * @since 3.1.0
 264 * @access private
 265 *
 266 * @param array $matches preg_replace_callback matches array
 267 * @return string
 268 */
 269function _autop_newline_preservation_helper( $matches ) {
 270	return str_replace("\n", "<WPPreserveNewline />", $matches[0]);
 271}
 272
 273/**
 274 * Don't auto-p wrap shortcodes that stand alone
 275 *
 276 * Ensures that shortcodes are not wrapped in <<p>>...<</p>>.
 277 *
 278 * @since 2.9.0
 279 *
 280 * @param string $pee The content.
 281 * @return string The filtered content.
 282 */
 283function shortcode_unautop( $pee ) {
 284	global $shortcode_tags;
 285
 286	if ( empty( $shortcode_tags ) || !is_array( $shortcode_tags ) ) {
 287		return $pee;
 288	}
 289
 290	$tagregexp = join( '|', array_map( 'preg_quote', array_keys( $shortcode_tags ) ) );
 291
 292	$pattern =
 293		  '/'
 294		. '<p>'                              // Opening paragraph
 295		. '\\s*+'                            // Optional leading whitespace
 296		. '('                                // 1: The shortcode
 297		.     '\\['                          // Opening bracket
 298		.     "($tagregexp)"                 // 2: Shortcode name
 299		.     '(?![\\w-])'                   // Not followed by word character or hyphen
 300		                                     // Unroll the loop: Inside the opening shortcode tag
 301		.     '[^\\]\\/]*'                   // Not a closing bracket or forward slash
 302		.     '(?:'
 303		.         '\\/(?!\\])'               // A forward slash not followed by a closing bracket
 304		.         '[^\\]\\/]*'               // Not a closing bracket or forward slash
 305		.     ')*?'
 306		.     '(?:'
 307		.         '\\/\\]'                   // Self closing tag and closing bracket
 308		.     '|'
 309		.         '\\]'                      // Closing bracket
 310		.         '(?:'                      // Unroll the loop: Optionally, anything between the opening and closing shortcode tags
 311		.             '[^\\[]*+'             // Not an opening bracket
 312		.             '(?:'
 313		.                 '\\[(?!\\/\\2\\])' // An opening bracket not followed by the closing shortcode tag
 314		.                 '[^\\[]*+'         // Not an opening bracket
 315		.             ')*+'
 316		.             '\\[\\/\\2\\]'         // Closing shortcode tag
 317		.         ')?'
 318		.     ')'
 319		. ')'
 320		. '\\s*+'                            // optional trailing whitespace
 321		. '<\\/p>'                           // closing paragraph
 322		. '/s';
 323
 324	return preg_replace( $pattern, '$1', $pee );
 325}
 326
 327/**
 328 * Checks to see if a string is utf8 encoded.
 329 *
 330 * NOTE: This function checks for 5-Byte sequences, UTF8
 331 *       has Bytes Sequences with a maximum length of 4.
 332 *
 333 * @author bmorel at ssi dot fr (modified)
 334 * @since 1.2.1
 335 *
 336 * @param string $str The string to be checked
 337 * @return bool True if $str fits a UTF-8 model, false otherwise.
 338 */
 339function seems_utf8($str) {
 340	$length = strlen($str);
 341	for ($i=0; $i < $length; $i++) {
 342		$c = ord($str[$i]);
 343		if ($c < 0x80) $n = 0; # 0bbbbbbb
 344		elseif (($c & 0xE0) == 0xC0) $n=1; # 110bbbbb
 345		elseif (($c & 0xF0) == 0xE0) $n=2; # 1110bbbb
 346		elseif (($c & 0xF8) == 0xF0) $n=3; # 11110bbb
 347		elseif (($c & 0xFC) == 0xF8) $n=4; # 111110bb
 348		elseif (($c & 0xFE) == 0xFC) $n=5; # 1111110b
 349		else return false; # Does not match any model
 350		for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
 351			if ((++$i == $length) || ((ord($str[$i]) & 0xC0) != 0x80))
 352				return false;
 353		}
 354	}
 355	return true;
 356}
 357
 358/**
 359 * Converts a number of special characters into their HTML entities.
 360 *
 361 * Specifically deals with: &, <, >, ", and '.
 362 *
 363 * $quote_style can be set to ENT_COMPAT to encode " to
 364 * &quot;, or ENT_QUOTES to do both. Default is ENT_NOQUOTES where no quotes are encoded.
 365 *
 366 * @since 1.2.2
 367 * @access private
 368 *
 369 * @param string $string The text which is to be encoded.
 370 * @param mixed $quote_style Optional. Converts double quotes if set to ENT_COMPAT, both single and double if set to ENT_QUOTES or none if set to ENT_NOQUOTES. Also compatible with old values; converting single quotes if set to 'single', double if set to 'double' or both if otherwise set. Default is ENT_NOQUOTES.
 371 * @param string $charset Optional. The character encoding of the string. Default is false.
 372 * @param boolean $double_encode Optional. Whether to encode existing html entities. Default is false.
 373 * @return string The encoded text with HTML entities.
 374 */
 375function _wp_specialchars( $string, $quote_style = ENT_NOQUOTES, $charset = false, $double_encode = false ) {
 376	$string = (string) $string;
 377
 378	if ( 0 === strlen( $string ) )
 379		return '';
 380
 381	// Don't bother if there are no specialchars - saves some processing
 382	if ( ! preg_match( '/[&<>"\']/', $string ) )
 383		return $string;
 384
 385	// Account for the previous behaviour of the function when the $quote_style is not an accepted value
 386	if ( empty( $quote_style ) )
 387		$quote_style = ENT_NOQUOTES;
 388	elseif ( ! in_array( $quote_style, array( 0, 2, 3, 'single', 'double' ), true ) )
 389		$quote_style = ENT_QUOTES;
 390
 391	// Store the site charset as a static to avoid multiple calls to wp_load_alloptions()
 392	if ( ! $charset ) {
 393		static $_charset;
 394		if ( ! isset( $_charset ) ) {
 395			$alloptions = wp_load_alloptions();
 396			$_charset = isset( $alloptions['blog_charset'] ) ? $alloptions['blog_charset'] : '';
 397		}
 398		$charset = $_charset;
 399	}
 400
 401	if ( in_array( $charset, array( 'utf8', 'utf-8', 'UTF8' ) ) )
 402		$charset = 'UTF-8';
 403
 404	$_quote_style = $quote_style;
 405
 406	if ( $quote_style === 'double' ) {
 407		$quote_style = ENT_COMPAT;
 408		$_quote_style = ENT_COMPAT;
 409	} elseif ( $quote_style === 'single' ) {
 410		$quote_style = ENT_NOQUOTES;
 411	}
 412
 413	// Handle double encoding ourselves
 414	if ( $double_encode ) {
 415		$string = @htmlspecialchars( $string, $quote_style, $charset );
 416	} else {
 417		// Decode &amp; into &
 418		$string = wp_specialchars_decode( $string, $_quote_style );
 419
 420		// Guarantee every &entity; is valid or re-encode the &
 421		$string = wp_kses_normalize_entities( $string );
 422
 423		// Now re-encode everything except &entity;
 424		$string = preg_split( '/(&#?x?[0-9a-z]+;)/i', $string, -1, PREG_SPLIT_DELIM_CAPTURE );
 425
 426		for ( $i = 0; $i < count( $string ); $i += 2 )
 427			$string[$i] = @htmlspecialchars( $string[$i], $quote_style, $charset );
 428
 429		$string = implode( '', $string );
 430	}
 431
 432	// Backwards compatibility
 433	if ( 'single' === $_quote_style )
 434		$string = str_replace( "'", '&#039;', $string );
 435
 436	return $string;
 437}
 438
 439/**
 440 * Converts a number of HTML entities into their special characters.
 441 *
 442 * Specifically deals with: &, <, >, ", and '.
 443 *
 444 * $quote_style can be set to ENT_COMPAT to decode " entities,
 445 * or ENT_QUOTES to do both " and '. Default is ENT_NOQUOTES where no quotes are decoded.
 446 *
 447 * @since 2.8.0
 448 *
 449 * @param string $string The text which is to be decoded.
 450 * @param mixed $quote_style Optional. Converts double quotes if set to ENT_COMPAT, both single and double if set to ENT_QUOTES or none if set to ENT_NOQUOTES. Also compatible with old _wp_specialchars() values; converting single quotes if set to 'single', double if set to 'double' or both if otherwise set. Default is ENT_NOQUOTES.
 451 * @return string The decoded text without HTML entities.
 452 */
 453function wp_specialchars_decode( $string, $quote_style = ENT_NOQUOTES ) {
 454	$string = (string) $string;
 455
 456	if ( 0 === strlen( $string ) ) {
 457		return '';
 458	}
 459
 460	// Don't bother if there are no entities - saves a lot of processing
 461	if ( strpos( $string, '&' ) === false ) {
 462		return $string;
 463	}
 464
 465	// Match the previous behaviour of _wp_specialchars() when the $quote_style is not an accepted value
 466	if ( empty( $quote_style ) ) {
 467		$quote_style = ENT_NOQUOTES;
 468	} elseif ( !in_array( $quote_style, array( 0, 2, 3, 'single', 'double' ), true ) ) {
 469		$quote_style = ENT_QUOTES;
 470	}
 471
 472	// More complete than get_html_translation_table( HTML_SPECIALCHARS )
 473	$single = array( '&#039;'  => '\'', '&#x27;' => '\'' );
 474	$single_preg = array( '/&#0*39;/'  => '&#039;', '/&#x0*27;/i' => '&#x27;' );
 475	$double = array( '&quot;' => '"', '&#034;'  => '"', '&#x22;' => '"' );
 476	$double_preg = array( '/&#0*34;/'  => '&#034;', '/&#x0*22;/i' => '&#x22;' );
 477	$others = array( '&lt;'   => '<', '&#060;'  => '<', '&gt;'   => '>', '&#062;'  => '>', '&amp;'  => '&', '&#038;'  => '&', '&#x26;' => '&' );
 478	$others_preg = array( '/&#0*60;/'  => '&#060;', '/&#0*62;/'  => '&#062;', '/&#0*38;/'  => '&#038;', '/&#x0*26;/i' => '&#x26;' );
 479
 480	if ( $quote_style === ENT_QUOTES ) {
 481		$translation = array_merge( $single, $double, $others );
 482		$translation_preg = array_merge( $single_preg, $double_preg, $others_preg );
 483	} elseif ( $quote_style === ENT_COMPAT || $quote_style === 'double' ) {
 484		$translation = array_merge( $double, $others );
 485		$translation_preg = array_merge( $double_preg, $others_preg );
 486	} elseif ( $quote_style === 'single' ) {
 487		$translation = array_merge( $single, $others );
 488		$translation_preg = array_merge( $single_preg, $others_preg );
 489	} elseif ( $quote_style === ENT_NOQUOTES ) {
 490		$translation = $others;
 491		$translation_preg = $others_preg;
 492	}
 493
 494	// Remove zero padding on numeric entities
 495	$string = preg_replace( array_keys( $translation_preg ), array_values( $translation_preg ), $string );
 496
 497	// Replace characters according to translation table
 498	return strtr( $string, $translation );
 499}
 500
 501/**
 502 * Checks for invalid UTF8 in a string.
 503 *
 504 * @since 2.8.0
 505 *
 506 * @param string $string The text which is to be checked.
 507 * @param boolean $strip Optional. Whether to attempt to strip out invalid UTF8. Default is false.
 508 * @return string The checked text.
 509 */
 510function wp_check_invalid_utf8( $string, $strip = false ) {
 511	$string = (string) $string;
 512
 513	if ( 0 === strlen( $string ) ) {
 514		return '';
 515	}
 516
 517	// Store the site charset as a static to avoid multiple calls to get_option()
 518	static $is_utf8;
 519	if ( !isset( $is_utf8 ) ) {
 520		$is_utf8 = in_array( get_option( 'blog_charset' ), array( 'utf8', 'utf-8', 'UTF8', 'UTF-8' ) );
 521	}
 522	if ( !$is_utf8 ) {
 523		return $string;
 524	}
 525
 526	// Check for support for utf8 in the installed PCRE library once and store the result in a static
 527	static $utf8_pcre;
 528	if ( !isset( $utf8_pcre ) ) {
 529		$utf8_pcre = @preg_match( '/^./u', 'a' );
 530	}
 531	// We can't demand utf8 in the PCRE installation, so just return the string in those cases
 532	if ( !$utf8_pcre ) {
 533		return $string;
 534	}
 535
 536	// preg_match fails when it encounters invalid UTF8 in $string
 537	if ( 1 === @preg_match( '/^./us', $string ) ) {
 538		return $string;
 539	}
 540
 541	// Attempt to strip the bad chars if requested (not recommended)
 542	if ( $strip && function_exists( 'iconv' ) ) {
 543		return iconv( 'utf-8', 'utf-8', $string );
 544	}
 545
 546	return '';
 547}
 548
 549/**
 550 * Encode the Unicode values to be used in the URI.
 551 *
 552 * @since 1.5.0
 553 *
 554 * @param string $utf8_string
 555 * @param int $length Max length of the string
 556 * @return string String with Unicode encoded for URI.
 557 */
 558function utf8_uri_encode( $utf8_string, $length = 0 ) {
 559	$unicode = '';
 560	$values = array();
 561	$num_octets = 1;
 562	$unicode_length = 0;
 563
 564	$string_length = strlen( $utf8_string );
 565	for ($i = 0; $i < $string_length; $i++ ) {
 566
 567		$value = ord( $utf8_string[ $i ] );
 568
 569		if ( $value < 128 ) {
 570			if ( $length && ( $unicode_length >= $length ) )
 571				break;
 572			$unicode .= chr($value);
 573			$unicode_length++;
 574		} else {
 575			if ( count( $values ) == 0 ) $num_octets = ( $value < 224 ) ? 2 : 3;
 576
 577			$values[] = $value;
 578
 579			if ( $length && ( $unicode_length + ($num_octets * 3) ) > $length )
 580				break;
 581			if ( count( $values ) == $num_octets ) {
 582				if ($num_octets == 3) {
 583					$unicode .= '%' . dechex($values[0]) . '%' . dechex($values[1]) . '%' . dechex($values[2]);
 584					$unicode_length += 9;
 585				} else {
 586					$unicode .= '%' . dechex($values[0]) . '%' . dechex($values[1]);
 587					$unicode_length += 6;
 588				}
 589
 590				$values = array();
 591				$num_octets = 1;
 592			}
 593		}
 594	}
 595
 596	return $unicode;
 597}
 598
 599/**
 600 * Converts all accent characters to ASCII characters.
 601 *
 602 * If there are no accent characters, then the string given is just returned.
 603 *
 604 * @since 1.2.1
 605 *
 606 * @param string $string Text that might have accent characters
 607 * @return string Filtered string with replaced "nice" characters.
 608 */
 609function remove_accents($string) {
 610	if ( !preg_match('/[\x80-\xff]/', $string) )
 611		return $string;
 612
 613	if (seems_utf8($string)) {
 614		$chars = array(
 615		// Decompositions for Latin-1 Supplement
 616		chr(194).chr(170) => 'a', chr(194).chr(186) => 'o',
 617		chr(195).chr(128) => 'A', chr(195).chr(129) => 'A',
 618		chr(195).chr(130) => 'A', chr(195).chr(131) => 'A',
 619		chr(195).chr(132) => 'A', chr(195).chr(133) => 'A',
 620		chr(195).chr(134) => 'AE',chr(195).chr(135) => 'C',
 621		chr(195).chr(136) => 'E', chr(195).chr(137) => 'E',
 622		chr(195).chr(138) => 'E', chr(195).chr(139) => 'E',
 623		chr(195).chr(140) => 'I', chr(195).chr(141) => 'I',
 624		chr(195).chr(142) => 'I', chr(195).chr(143) => 'I',
 625		chr(195).chr(144) => 'D', chr(195).chr(145) => 'N',
 626		chr(195).chr(146) => 'O', chr(195).chr(147) => 'O',
 627		chr(195).chr(148) => 'O', chr(195).chr(149) => 'O',
 628		chr(195).chr(150) => 'O', chr(195).chr(153) => 'U',
 629		chr(195).chr(154) => 'U', chr(195).chr(155) => 'U',
 630		chr(195).chr(156) => 'U', chr(195).chr(157) => 'Y',
 631		chr(195).chr(158) => 'TH',chr(195).chr(159) => 's',
 632		chr(195).chr(160) => 'a', chr(195).chr(161) => 'a',
 633		chr(195).chr(162) => 'a', chr(195).chr(163) => 'a',
 634		chr(195).chr(164) => 'a', chr(195).chr(165) => 'a',
 635		chr(195).chr(166) => 'ae',chr(195).chr(167) => 'c',
 636		chr(195).chr(168) => 'e', chr(195).chr(169) => 'e',
 637		chr(195).chr(170) => 'e', chr(195).chr(171) => 'e',
 638		chr(195).chr(172) => 'i', chr(195).chr(173) => 'i',
 639		chr(195).chr(174) => 'i', chr(195).chr(175) => 'i',
 640		chr(195).chr(176) => 'd', chr(195).chr(177) => 'n',
 641		chr(195).chr(178) => 'o', chr(195).chr(179) => 'o',
 642		chr(195).chr(180) => 'o', chr(195).chr(181) => 'o',
 643		chr(195).chr(182) => 'o', chr(195).chr(184) => 'o',
 644		chr(195).chr(185) => 'u', chr(195).chr(186) => 'u',
 645		chr(195).chr(187) => 'u', chr(195).chr(188) => 'u',
 646		chr(195).chr(189) => 'y', chr(195).chr(190) => 'th',
 647		chr(195).chr(191) => 'y', chr(195).chr(152) => 'O',
 648		// Decompositions for Latin Extended-A
 649		chr(196).chr(128) => 'A', chr(196).chr(129) => 'a',
 650		chr(196).chr(130) => 'A', chr(196).chr(131) => 'a',
 651		chr(196).chr(132) => 'A', chr(196).chr(133) => 'a',
 652		chr(196).chr(134) => 'C', chr(196).chr(135) => 'c',
 653		chr(196).chr(136) => 'C', chr(196).chr(137) => 'c',
 654		chr(196).chr(138) => 'C', chr(196).chr(139) => 'c',
 655		chr(196).chr(140) => 'C', chr(196).chr(141) => 'c',
 656		chr(196).chr(142) => 'D', chr(196).chr(143) => 'd',
 657		chr(196).chr(144) => 'D', chr(196).chr(145) => 'd',
 658		chr(196).chr(146) => 'E', chr(196).chr(147) => 'e',
 659		chr(196).chr(148) => 'E', chr(196).chr(149) => 'e',
 660		chr(196).chr(150) => 'E', chr(196).chr(151) => 'e',
 661		chr(196).chr(152) => 'E', chr(196).chr(153) => 'e',
 662		chr(196).chr(154) => 'E', chr(196).chr(155) => 'e',
 663		chr(196).chr(156) => 'G', chr(196).chr(157) => 'g',
 664		chr(196).chr(158) => 'G', chr(196).chr(159) => 'g',
 665		chr(196).chr(160) => 'G', chr(196).chr(161) => 'g',
 666		chr(196).chr(162) => 'G', chr(196).chr(163) => 'g',
 667		chr(196).chr(164) => 'H', chr(196).chr(165) => 'h',
 668		chr(196).chr(166) => 'H', chr(196).chr(167) => 'h',
 669		chr(196).chr(168) => 'I', chr(196).chr(169) => 'i',
 670		chr(196).chr(170) => 'I', chr(196).chr(171) => 'i',
 671		chr(196).chr(172) => 'I', chr(196).chr(173) => 'i',
 672		chr(196).chr(174) => 'I', chr(196).chr(175) => 'i',
 673		chr(196).chr(176) => 'I', chr(196).chr(177) => 'i',
 674		chr(196).chr(178) => 'IJ',chr(196).chr(179) => 'ij',
 675		chr(196).chr(180) => 'J', chr(196).chr(181) => 'j',
 676		chr(196).chr(182) => 'K', chr(196).chr(183) => 'k',
 677		chr(196).chr(184) => 'k', chr(196).chr(185) => 'L',
 678		chr(196).chr(186) => 'l', chr(196).chr(187) => 'L',
 679		chr(196).chr(188) => 'l', chr(196).chr(189) => 'L',
 680		chr(196).chr(190) => 'l', chr(196).chr(191) => 'L',
 681		chr(197).chr(128) => 'l', chr(197).chr(129) => 'L',
 682		chr(197).chr(130) => 'l', chr(197).chr(131) => 'N',
 683		chr(197).chr(132) => 'n', chr(197).chr(133) => 'N',
 684		chr(197).chr(134) => 'n', chr(197).chr(135) => 'N',
 685		chr(197).chr(136) => 'n', chr(197).chr(137) => 'N',
 686		chr(197).chr(138) => 'n', chr(197).chr(139) => 'N',
 687		chr(197).chr(140) => 'O', chr(197).chr(141) => 'o',
 688		chr(197).chr(142) => 'O', chr(197).chr(143) => 'o',
 689		chr(197).chr(144) => 'O', chr(197).chr(145) => 'o',
 690		chr(197).chr(146) => 'OE',chr(197).chr(147) => 'oe',
 691		chr(197).chr(148) => 'R',chr(197).chr(149) => 'r',
 692		chr(197).chr(150) => 'R',chr(197).chr(151) => 'r',
 693		chr(197).chr(152) => 'R',chr(197).chr(153) => 'r',
 694		chr(197).chr(154) => 'S',chr(197).chr(155) => 's',
 695		chr(197).chr(156) => 'S',chr(197).chr(157) => 's',
 696		chr(197).chr(158) => 'S',chr(197).chr(159) => 's',
 697		chr(197).chr(160) => 'S', chr(197).chr(161) => 's',
 698		chr(197).chr(162) => 'T', chr(197).chr(163) => 't',
 699		chr(197).chr(164) => 'T', chr(197).chr(165) => 't',
 700		chr(197).chr(166) => 'T', chr(197).chr(167) => 't',
 701		chr(197).chr(168) => 'U', chr(197).chr(169) => 'u',
 702		chr(197).chr(170) => 'U', chr(197).chr(171) => 'u',
 703		chr(197).chr(172) => 'U', chr(197).chr(173) => 'u',
 704		chr(197).chr(174) => 'U', chr(197).chr(175) => 'u',
 705		chr(197).chr(176) => 'U', chr(197).chr(177) => 'u',
 706		chr(197).chr(178) => 'U', chr(197).chr(179) => 'u',
 707		chr(197).chr(180) => 'W', chr(197).chr(181) => 'w',
 708		chr(197).chr(182) => 'Y', chr(197).chr(183) => 'y',
 709		chr(197).chr(184) => 'Y', chr(197).chr(185) => 'Z',
 710		chr(197).chr(186) => 'z', chr(197).chr(187) => 'Z',
 711		chr(197).chr(188) => 'z', chr(197).chr(189) => 'Z',
 712		chr(197).chr(190) => 'z', chr(197).chr(191) => 's',
 713		// Decompositions for Latin Extended-B
 714		chr(200).chr(152) => 'S', chr(200).chr(153) => 's',
 715		chr(200).chr(154) => 'T', chr(200).chr(155) => 't',
 716		// Euro Sign
 717		chr(226).chr(130).chr(172) => 'E',
 718		// GBP (Pound) Sign
 719		chr(194).chr(163) => '',
 720		// Vowels with diacritic (Vietnamese)
 721		// unmarked
 722		chr(198).chr(160) => 'O', chr(198).chr(161) => 'o',
 723		chr(198).chr(175) => 'U', chr(198).chr(176) => 'u',
 724		// grave accent
 725		chr(225).chr(186).chr(166) => 'A', chr(225).chr(186).chr(167) => 'a',
 726		chr(225).chr(186).chr(176) => 'A', chr(225).chr(186).chr(177) => 'a',
 727		chr(225).chr(187).chr(128) => 'E', chr(225).chr(187).chr(129) => 'e',
 728		chr(225).chr(187).chr(146) => 'O', chr(225).chr(187).chr(147) => 'o',
 729		chr(225).chr(187).chr(156) => 'O', chr(225).chr(187).chr(157) => 'o',
 730		chr(225).chr(187).chr(170) => 'U', chr(225).chr(187).chr(171) => 'u',
 731		chr(225).chr(187).chr(178) => 'Y', chr(225).chr(187).chr(179) => 'y',
 732		// hook
 733		chr(225).chr(186).chr(162) => 'A', chr(225).chr(186).chr(163) => 'a',
 734		chr(225).chr(186).chr(168) => 'A', chr(225).chr(186).chr(169) => 'a',
 735		chr(225).chr(186).chr(178) => 'A', chr(225).chr(186).chr(179) => 'a',
 736		chr(225).chr(186).chr(186) => 'E', chr(225).chr(186).chr(187) => 'e',
 737		chr(225).chr(187).chr(130) => 'E', chr(225).chr(187).chr(131) => 'e',
 738		chr(225).chr(187).chr(136) => 'I', chr(225).chr(187).chr(137) => 'i',
 739		chr(225).chr(187).chr(142) => 'O', chr(225).chr(187).chr(143) => 'o',
 740		chr(225).chr(187).chr(148) => 'O', chr(225).chr(187).chr(149) => 'o',
 741		chr(225).chr(187).chr(158) => 'O', chr(225).chr(187).chr(159) => 'o',
 742		chr(225).chr(187).chr(166) => 'U', chr(225).chr(187).chr(167) => 'u',
 743		chr(225).chr(187).chr(172) => 'U', chr(225).chr(187).chr(173) => 'u',
 744		chr(225).chr(187).chr(182) => 'Y', chr(225).chr(187).chr(183) => 'y',
 745		// tilde
 746		chr(225).chr(186).chr(170) => 'A', chr(225).chr(186).chr(171) => 'a',
 747		chr(225).chr(186).chr(180) => 'A', chr(225).chr(186).chr(181) => 'a',
 748		chr(225).chr(186).chr(188) => 'E', chr(225).chr(186).chr(189) => 'e',
 749		chr(225).chr(187).chr(132) => 'E', chr(225).chr(187).chr(133) => 'e',
 750		chr(225).chr(187).chr(150) => 'O', chr(225).chr(187).chr(151) => 'o',
 751		chr(225).chr(187).chr(160) => 'O', chr(225).chr(187).chr(161) => 'o',
 752		chr(225).chr(187).chr(174) => 'U', chr(225).chr(187).chr(175) => 'u',
 753		chr(225).chr(187).chr(184) => 'Y', chr(225).chr(187).chr(185) => 'y',
 754		// acute accent
 755		chr(225).chr(186).chr(164) => 'A', chr(225).chr(186).chr(165) => 'a',
 756		chr(225).chr(186).chr(174) => 'A', chr(225).chr(186).chr(175) => 'a',
 757		chr(225).chr(186).chr(190) => 'E', chr(225).chr(186).chr(191) => 'e',
 758		chr(225).chr(187).chr(144) => 'O', chr(225).chr(187).chr(145) => 'o',
 759		chr(225).chr(187).chr(154) => 'O', chr(225).chr(187).chr(155) => 'o',
 760		chr(225).chr(187).chr(168) => 'U', chr(225).chr(187).chr(169) => 'u',
 761		// dot below
 762		chr(225).chr(186).chr(160) => 'A', chr(225).chr(186).chr(161) => 'a',
 763		chr(225).chr(186).chr(172) => 'A', chr(225).chr(186).chr(173) => 'a',
 764		chr(225).chr(186).chr(182) => 'A', chr(225).chr(186).chr(183) => 'a',
 765		chr(225).chr(186).chr(184) => 'E', chr(225).chr(186).chr(185) => 'e',
 766		chr(225).chr(187).chr(134) => 'E', chr(225).chr(187).chr(135) => 'e',
 767		chr(225).chr(187).chr(138) => 'I', chr(225).chr(187).chr(139) => 'i',
 768		chr(225).chr(187).chr(140) => 'O', chr(225).chr(187).chr(141) => 'o',
 769		chr(225).chr(187).chr(152) => 'O', chr(225).chr(187).chr(153) => 'o',
 770		chr(225).chr(187).chr(162) => 'O', chr(225).chr(187).chr(163) => 'o',
 771		chr(225).chr(187).chr(164) => 'U', chr(225).chr(187).chr(165) => 'u',
 772		chr(225).chr(187).chr(176) => 'U', chr(225).chr(187).chr(177) => 'u',
 773		chr(225).chr(187).chr(180) => 'Y', chr(225).chr(187).chr(181) => 'y',
 774		// Vowels with diacritic (Chinese, Hanyu Pinyin)
 775		chr(201).chr(145) => 'a',
 776		// macron
 777		chr(199).chr(149) => 'U', chr(199).chr(150) => 'u',
 778		// acute accent
 779		chr(199).chr(151) => 'U', chr(199).chr(152) => 'u',
 780		// caron
 781		chr(199).chr(141) => 'A', chr(199).chr(142) => 'a',
 782		chr(199).chr(143) => 'I', chr(199).chr(144) => 'i',
 783		chr(199).chr(145) => 'O', chr(199).chr(146) => 'o',
 784		chr(199).chr(147) => 'U', chr(199).chr(148) => 'u',
 785		chr(199).chr(153) => 'U', chr(199).chr(154) => 'u',
 786		// grave accent
 787		chr(199).chr(155) => 'U', chr(199).chr(156) => 'u',
 788		);
 789
 790		// Used for locale-specific rules
 791		$locale = get_locale();
 792
 793		if ( 'de_DE' == $locale ) {
 794			$chars[ chr(195).chr(132) ] = 'Ae';
 795			$chars[ chr(195).chr(164) ] = 'ae';
 796			$chars[ chr(195).chr(150) ] = 'Oe';
 797			$chars[ chr(195).chr(182) ] = 'oe';
 798			$chars[ chr(195).chr(156) ] = 'Ue';
 799			$chars[ chr(195).chr(188) ] = 'ue';
 800			$chars[ chr(195).chr(159) ] = 'ss';
 801		}
 802
 803		$string = strtr($string, $chars);
 804	} else {
 805		// Assume ISO-8859-1 if not UTF-8
 806		$chars['in'] = chr(128).chr(131).chr(138).chr(142).chr(154).chr(158)
 807			.chr(159).chr(162).chr(165).chr(181).chr(192).chr(193).chr(194)
 808			.chr(195).chr(196).chr(197).chr(199).chr(200).chr(201).chr(202)
 809			.chr(203).chr(204).chr(205).chr(206).chr(207).chr(209).chr(210)
 810			.chr(211).chr(212).chr(213).chr(214).chr(216).chr(217).chr(218)
 811			.chr(219).chr(220).chr(221).chr(224).chr(225).chr(226).chr(227)
 812			.chr(228).chr(229).chr(231).chr(232).chr(233).chr(234).chr(235)
 813			.chr(236).chr(237).chr(238).chr(239).chr(241).chr(242).chr(243)
 814			.chr(244).chr(245).chr(246).chr(248).chr(249).chr(250).chr(251)
 815			.chr(252).chr(253).chr(255);
 816
 817		$chars['out'] = "EfSZszYcYuAAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuyy";
 818
 819		$string = strtr($string, $chars['in'], $chars['out']);
 820		$double_chars['in'] = array(chr(140), chr(156), chr(198), chr(208), chr(222), chr(223), chr(230), chr(240), chr(254));
 821		$double_chars['out'] = array('OE', 'oe', 'AE', 'DH', 'TH', 'ss', 'ae', 'dh', 'th');
 822		$string = str_replace($double_chars['in'], $double_chars['out'], $string);
 823	}
 824
 825	return $string;
 826}
 827
 828/**
 829 * Sanitizes a filename, replacing whitespace with dashes.
 830 *
 831 * Removes special characters that are illegal in filenames on certain
 832 * operating systems and special characters requiring special escaping
 833 * to manipulate at the command line. Replaces spaces and consecutive
 834 * dashes with a single dash. Trims period, dash and underscore from beginning
 835 * and end of filename.
 836 *
 837 * @since 2.1.0
 838 *
 839 * @param string $filename The filename to be sanitized
 840 * @return string The sanitized filename
 841 */
 842function sanitize_file_name( $filename ) {
 843	$filename_raw = $filename;
 844	$special_chars = array("?", "[", "]", "/", "\\", "=", "<", ">", ":", ";", ",", "'", "\"", "&", "$", "#", "*", "(", ")", "|", "~", "`", "!", "{", "}", chr(0));
 845	$special_chars = apply_filters('sanitize_file_name_chars', $special_chars, $filename_raw);
 846	$filename = str_replace($special_chars, '', $filename);
 847	$filename = preg_replace('/[\s-]+/', '-', $filename);
 848	$filename = trim($filename, '.-_');
 849
 850	// Split the filename into a base and extension[s]
 851	$parts = explode('.', $filename);
 852
 853	// Return if only one extension
 854	if ( count($parts) <= 2 )
 855		return apply_filters('sanitize_file_name', $filename, $filename_raw);
 856
 857	// Process multiple extensions
 858	$filename = array_shift($parts);
 859	$extension = array_pop($parts);
 860	$mimes = get_allowed_mime_types();
 861
 862	// Loop over any intermediate extensions. Munge them with a trailing underscore if they are a 2 - 5 character
 863	// long alpha string not in the extension whitelist.
 864	foreach ( (array) $parts as $part) {
 865		$filename .= '.' . $part;
 866
 867		if ( preg_match("/^[a-zA-Z]{2,5}\d?$/", $part) ) {
 868			$allowed = false;
 869			foreach ( $mimes as $ext_preg => $mime_match ) {
 870				$ext_preg = '!^(' . $ext_preg . ')$!i';
 871				if ( preg_match( $ext_preg, $part ) ) {
 872					$allowed = true;
 873					break;
 874				}
 875			}
 876			if ( !$allowed )
 877				$filename .= '_';
 878		}
 879	}
 880	$filename .= '.' . $extension;
 881
 882	return apply_filters('sanitize_file_name', $filename, $filename_raw);
 883}
 884
 885/**
 886 * Sanitizes a username, stripping out unsafe characters.
 887 *
 888 * Removes tags, octets, entities, and if strict is enabled, will only keep
 889 * alphanumeric, _, space, ., -, @. After sanitizing, it passes the username,
 890 * raw username (the username in the parameter), and the value of $strict as
 891 * parameters for the 'sanitize_user' filter.
 892 *
 893 * @since 2.0.0
 894 * @uses apply_filters() Calls 'sanitize_user' hook on username, raw username,
 895 *		and $strict parameter.
 896 *
 897 * @param string $username The username to be sanitized.
 898 * @param bool $strict If set limits $username to specific characters. Default false.
 899 * @return string The sanitized username, after passing through filters.
 900 */
 901function sanitize_user( $username, $strict = false ) {
 902	$raw_username = $username;
 903	$username = wp_strip_all_tags( $username );
 904	$username = remove_accents( $username );
 905	// Kill octets
 906	$username = preg_replace( '|%([a-fA-F0-9][a-fA-F0-9])|', '', $username );
 907	$username = preg_replace( '/&.+?;/', '', $username ); // Kill entities
 908
 909	// If strict, reduce to ASCII for max portability.
 910	if ( $strict )
 911		$username = preg_replace( '|[^a-z0-9 _.\-@]|i', '', $username );
 912
 913	$username = trim( $username );
 914	// Consolidate contiguous whitespace
 915	$username = preg_replace( '|\s+|', ' ', $username );
 916
 917	return apply_filters( 'sanitize_user', $username, $raw_username, $strict );
 918}
 919
 920/**
 921 * Sanitizes a string key.
 922 *
 923 * Keys are used as internal identifiers. Lowercase alphanumeric characters, dashes and underscores are allowed.
 924 *
 925 * @since 3.0.0
 926 *
 927 * @param string $key String key
 928 * @return string Sanitized key
 929 */
 930function sanitize_key( $key ) {
 931	$raw_key = $key;
 932	$key = strtolower( $key );
 933	$key = preg_replace( '/[^a-z0-9_\-]/', '', $key );
 934	return apply_filters( 'sanitize_key', $key, $raw_key );
 935}
 936
 937/**
 938 * Sanitizes a title, or returns a fallback title.
 939 *
 940 * Specifically, HTML and PHP tags are stripped. Further actions can be added
 941 * via the plugin API. If $title is empty and $fallback_title is set, the latter
 942 * will be used.
 943 *
 944 * @since 1.0.0
 945 *
 946 * @param string $title The string to be sanitized.
 947 * @param string $fallback_title Optional. A title to use if $title is empty.
 948 * @param string $context Optional. The operation for which the string is sanitized
 949 * @return string The sanitized string.
 950 */
 951function sanitize_title( $title, $fallback_title = '', $context = 'save' ) {
 952	$raw_title = $title;
 953
 954	if ( 'save' == $context )
 955		$title = remove_accents($title);
 956
 957	$title = apply_filters('sanitize_title', $title, $raw_title, $context);
 958
 959	if ( '' === $title || false === $title )
 960		$title = $fallback_title;
 961
 962	return $title;
 963}
 964
 965/**
 966 * Sanitizes a title with the 'query' context.
 967 *
 968 * Used for querying the database for a value from URL.
 969 *
 970 * @since 3.1.0
 971 * @uses sanitize_title()
 972 *
 973 * @param string $title The string to be sanitized.
 974 * @return string The sanitized string.
 975 */
 976function sanitize_title_for_query( $title ) {
 977	return sanitize_title( $title, '', 'query' );
 978}
 979
 980/**
 981 * Sanitizes a title, replacing whitespace and a few other characters with dashes.
 982 *
 983 * Limits the output to alphanumeric characters, underscore (_) and dash (-).
 984 * Whitespace becomes a dash.
 985 *
 986 * @since 1.2.0
 987 *
 988 * @param string $title The title to be sanitized.
 989 * @param string $raw_title Optional. Not used.
 990 * @param string $context Optional. The operation for which the string is sanitized.
 991 * @return string The sanitized title.
 992 */
 993function sanitize_title_with_dashes( $title, $raw_title = '', $context = 'display' ) {
 994	$title = strip_tags($title);
 995	// Preserve escaped octets.
 996	$title = preg_replace('|%([a-fA-F0-9][a-fA-F0-9])|', '---$1---', $title);
 997	// Remove percent signs that are not part of an octet.
 998	$title = str_replace('%', '', $title);
 999	// Restore octets.
1000	$title = preg_replace('|---([a-fA-F0-9][a-fA-F0-9])---|', '%$1', $title);
1001
1002	if (seems_utf8($title)) {
1003		if (function_exists('mb_strtolower')) {
1004			$title = mb_strtolower($title, 'UTF-8');
1005		}
1006		$title = utf8_uri_encode($title, 200);
1007	}
1008
1009	$title = strtolower($title);
1010	$title = preg_replace('/&.+?;/', '', $title); // kill entities
1011	$title = str_replace('.', '-', $title);
1012
1013	if ( 'save' == $context ) {
1014		// Convert nbsp, ndash and mdash to hyphens
1015		$title = str_replace( array( '%c2%a0', '%e2%80%93', '%e2%80%94' ), '-', $title );
1016
1017		// Strip these characters entirely
1018		$title = str_replace( array(
1019			// iexcl and iquest
1020			'%c2%a1', '%c2%bf',
1021			// angle quotes
1022			'%c2%ab', '%c2%bb', '%e2%80%b9', '%e2%80%ba',
1023			// curly quotes
1024			'%e2%80%98', '%e2%80%99', '%e2%80%9c', '%e2%80%9d',
1025			'%e2%80%9a', '%e2%80%9b', '%e2%80%9e', '%e2%80%9f',
1026			// copy, reg, deg, hellip and trade
1027			'%c2%a9', '%c2%ae', '%c2%b0', '%e2%80%a6', '%e2%84%a2',
1028			// acute accents
1029			'%c2%b4', '%cb%8a', '%cc%81', '%cd%81',
1030			// grave accent, macron, caron
1031			'%cc%80', '%cc%84', '%cc%8c',
1032		), '', $title );
1033
1034		// Convert times to x
1035		$title = str_replace( '%c3%97', 'x', $title );
1036	}
1037
1038	$title = preg_replace('/[^%a-z0-9 _-]/', '', $title);
1039	$title = preg_replace('/\s+/', '-', $title);
1040	$title = preg_replace('|-+|', '-', $title);
1041	$title = trim($title, '-');
1042
1043	return $title;
1044}
1045
1046/**
1047 * Ensures a string is a valid SQL order by clause.
1048 *
1049 * Accepts one or more columns, with or without ASC/DESC, and also accepts
1050 * RAND().
1051 *
1052 * @since 2.5.1
1053 *
1054 * @param string $orderby Order by string to be checked.
1055 * @return string|bool Returns the order by clause if it is a match, false otherwise.
1056 */
1057function sanitize_sql_orderby( $orderby ){
1058	preg_match('/^\s*([a-z0-9_]+(\s+(ASC|DESC))?(\s*,\s*|\s*$))+|^\s*RAND\(\s*\)\s*$/i', $orderby, $obmatches);
1059	if ( !$obmatches )
1060		return false;
1061	return $orderby;
1062}
1063
1064/**
1065 * Sanitizes an HTML classname to ensure it only contains valid characters.
1066 *
1067 * Strips the string down to A-Z,a-z,0-9,_,-. If this results in an empty
1068 * string then it will return the alternative value supplied.
1069 *
1070 * @todo Expand to support the full range of CDATA that a class attribute can contain.
1071 *
1072 * @since 2.8.0
1073 *
1074 * @param string $class The classname to be sanitized
1075 * @param string $fallback Optional. The value to return if the sanitization end's up as an empty string.
1076 * 	Defaults to an empty string.
1077 * @return string The sanitized value
1078 */
1079function sanitize_html_class( $class, $fallback = '' ) {
1080	//Strip out any % encoded octets
1081	$sanitized = preg_replace( '|%[a-fA-F0-9][a-fA-F0-9]|', '', $class );
1082
1083	//Limit to A-Z,a-z,0-9,_,-
1084	$sanitized = preg_replace( '/[^A-Za-z0-9_-]/', '', $sanitized );
1085
1086	if ( '' == $sanitized )
1087		$sanitized = $fallback;
1088
1089	return apply_filters( 'sanitize_html_class', $sanitized, $class, $fallback );
1090}
1091
1092/**
1093 * Converts a number of characters from a string.
1094 *
1095 * Metadata tags <<title>> and <<category>> are removed, <<br>> and <<hr>> are
1096 * converted into correct XHTML and Unicode characters are converted to the
1097 * valid range.
1098 *
1099 * @since 0.71
1100 *
1101 * @param string $content String of characters to be converted.
1102 * @param string $deprecated Not used.
1103 * @return string Converted string.
1104 */
1105function convert_chars($content, $deprecated = '') {
1106	if ( !empty( $deprecated ) )
1107		_deprecated_argument( __FUNCTION__, '0.71' );
1108
1109	// Translation of invalid Unicode references range to valid range
1110	$wp_htmltranswinuni = array(
1111	'&#128;' => '&#8364;', // the Euro sign
1112	'&#129;' => '',
1113	'&#130;' => '&#8218;', // these are Windows CP1252 specific characters
1114	'&#131;' => '&#402;',  // they would look weird on non-Windows browsers
1115	'&#132;' => '&#8222;',
1116	'&#133;' => '&#8230;',
1117	'&#134;' => '&#8224;',
1118	'&#135;' => '&#8225;',
1119	'&#136;' => '&#710;',
1120	'&#137;' => '&#8240;',
1121	'&#138;' => '&#352;',
1122	'&#139;' => '&#8249;',
1123	'&#140;' => '&#338;',
1124	'&#141;' => '',
1125	'&#142;' => '&#381;',
1126	'&#143;' => '',
1127	'&#144;' => '',
1128	'&#145;' => '&#8216;',
1129	'&#146;' => '&#8217;',
1130	'&#147;' => '&#8220;',
1131	'&#148;' => '&#8221;',
1132	'&#149;' => '&#8226;',
1133	'&#150;' => '&#8211;',
1134	'&#151;' => '&#8212;',
1135	'&#152;' => '&#732;',
1136	'&#153;' => '&#8482;',
1137	'&#154;' => '&#353;',
1138	'&#155;' => '&#8250;',
1139	'&#156;' => '&#339;',
1140	'&#157;' => '',
1141	'&#158;' => '&#382;',
1142	'&#159;' => '&#376;'
1143	);
1144
1145	// Remove metadata tags
1146	$content = preg_replace('/<title>(.+?)<\/title>/','',$content);
1147	$content = preg_replace('/<category>(.+?)<\/category>/','',$content);
1148
1149	// Converts lone & characters into &#38; (a.k.a. &amp;)
1150	$content = preg_replace('/&([^#])(?![a-z1-4]{1,8};)/i', '&#038;$1', $content);
1151
1152	// Fix Word pasting
1153	$content = strtr($content, $wp_htmltranswinuni);
1154
1155	// Just a little XHTML help
1156	$content = str_replace('<br>', '<br />', $content);
1157	$content = str_replace('<hr>', '<hr />', $content);
1158
1159	return $content;
1160}
1161
1162/**
1163 * Balances tags if forced to, or if the 'use_balanceTags' option is set to true.
1164 *
1165 * @since 0.71
1166 *
1167 * @param string $text Text to be balanced
1168 * @param bool $force If true, forces balancing, ignoring the value of the option. Default false.
1169 * @return string Balanced text
1170 */
1171function balanceTags( $text, $force = false ) {
1172	if ( $force || get_option('use_balanceTags') == 1 )
1173		return force_balance_tags( $text );
1174	else
1175		return $text;
1176}
1177
1178/**
1179 * Balances tags of string using a modified stack.
1180 *
1181 * @since 2.0.4
1182 *
1183 * @author Leonard Lin <leonard@acm.org>
1184 * @license GPL
1185 * @copyright November 4, 2001
1186 * @version 1.1
1187 * @todo Make better - change loop condition to $text in 1.2
1188 * @internal Modified by Scott Reilly (coffee2code) 02 Aug 2004
1189 *		1.1  Fixed handling of append/stack pop order of end text
1190 *			 Added Cleaning Hooks
1191 *		1.0  First Version
1192 *
1193 * @param string $text Text to be balanced.
1194 * @return string Balanced text.
1195 */
1196function force_balance_tags( $text ) {
1197	$tagstack = array();
1198	$stacksize = 0;
1199	$tagqueue = '';
1200	$newtext = '';
1201	// Known single-entity/self-closing tags
1202	$single_tags = array( 'area', 'base', 'basefont', 'br', 'col', 'command', 'embed', 'frame', 'hr', 'img', 'input', 'isindex', 'link', 'meta', 'param', 'source' );
1203	// Tags that can be immediately nested within themselves
1204	$nestable_tags = array( 'blockquote', 'div', 'object', 'q', 'span' );
1205
1206	// WP bug fix for comments - in case you REALLY meant to type '< !--'
1207	$text = str_replace('< !--', '<    !--', $text);
1208	// WP bug fix for LOVE <3 (and other situations with '<' before a number)
1209	$text = preg_replace('#<([0-9]{1})#', '&lt;$1', $text);
1210
1211	while ( preg_match("/<(\/?[\w:]*)\s*([^>]*)>/", $text, $regex) ) {
1212		$newtext .= $tagqueue;
1213
1214		$i = strpos($text, $regex[0]);
1215		$l = strlen($regex[0]);
1216
1217		// clear the shifter
1218		$tagqueue = '';
1219		// Pop or Push
1220		if ( isset($regex[1][0]) && '/' == $regex[1][0] ) { // End Tag
1221			$tag = strtolower(substr($regex[1],1));
1222			// if too many closing tags
1223			if( $stacksize <= 0 ) {
1224				$tag = '';
1225				// or close to be safe $tag = '/' . $tag;
1226			}
1227			// if stacktop value = tag close value then pop
1228			else if ( $tagstack[$stacksize - 1] == $tag ) { // found closing tag
1229				$tag = '</' . $tag . '>'; // Close Tag
1230				// Pop
1231				array_pop( $tagstack );
1232				$stacksize--;
1233			} else { // closing tag not at top, search for it
1234				for ( $j = $stacksize-1; $j >= 0; $j-- ) {
1235					if ( $tagstack[$j] == $tag ) {
1236					// add tag to tagqueue
1237						for ( $k = $stacksize-1; $k >= $j; $k--) {
1238							$tagqueue .= '</' . array_pop( $tagstack ) . '>';
1239							$stacksize--;
1240						}
1241						break;
1242					}
1243				}
1244				$tag = '';
1245			}
1246		} else { // Begin Tag
1247			$tag = strtolower($regex[1]);
1248
1249			// Tag Cleaning
1250
1251			// If it's an empty tag "< >", do nothing
1252			if ( '' == $tag ) {
1253				// do nothing
1254			}
1255			// ElseIf it presents itself as a self-closing tag...
1256			elseif ( substr( $regex[2], -1 ) == '/' ) {
1257				// ...but it isn't a known single-entity self-closing tag, then don't let it be treated as such and
1258				// immediately close it with a closing tag (the tag will encapsulate no text as a result)
1259				if ( ! in_array( $tag, $single_tags ) )
1260					$regex[2] = trim( substr( $regex[2], 0, -1 ) ) . "></$tag";
1261			}
1262			// ElseIf it's a known single-entity tag but it doesn't close itself, do so
1263			elseif ( in_array($tag, $single_tags) ) {
1264				$regex[2] .= '/';
1265			}
1266			// Else it's not a single-entity tag
1267			else {
1268				// If the top of the stack is the same as the tag we want to push, close previous tag
1269				if ( $stacksize > 0 && !in_array($tag, $nestable_tags) && $tagstack[$stacksize - 1] == $tag ) {
1270					$tagqueue = '</' . array_pop( $tagstack ) . '>';
1271					$stacksize--;
1272				}
1273				$stacksize = array_push( $tagstack, $tag );
1274			}
1275
1276			// Attributes
1277			$attributes = $regex[2];
1278			if( ! empty( $attributes ) && $attributes[0] != '>' )
1279				$attributes = ' ' . $attributes;
1280
1281			$tag = '<' . $tag . $attributes . '>';
1282			//If already queuing a close tag, then put this tag on, too
1283			if ( !empty($tagqueue) ) {
1284				$tagqueue .= $tag;
1285				$tag = '';
1286			}
1287		}
1288		$newtext .= substr($text, 0, $i) . $tag;
1289		$text = substr($text, $i + $l);
1290	}
1291
1292	// Clear Tag Queue
1293	$newtext .= $tagqueue;
1294
1295	// Add Remaining text
1296	$newtext .= $text;
1297
1298	// Empty Stack
1299	while( $x = array_pop($tagstack) )
1300		$newtext .= '</' . $x . '>'; // Add remaining tags to close
1301
1302	// WP fix for the bug with HTML comments
1303	$newtext = str_replace("< !--","<!--",$newtext);
1304	$newtext = str_replace("<    !--","< !--",$newtext);
1305
1306	return $newtext;
1307}
1308
1309/**
1310 * Acts on text which is about to be edited.
1311 *
1312 * The $content is run through esc_textarea(), which uses htmlspecialchars()
1313 * to convert special characters to HTML entities. If $richedit is set to true,
1314 * it is simply a holder for the 'format_to_edit' filter.
1315 *
1316 * @since 0.71
1317 *
1318 * @param string $content The text about to be edited.
1319 * @param bool $richedit Whether the $content should not pass through htmlspecialchars(). Default false (meaning it will be passed).
1320 * @return string The text after the filter (and possibly htmlspecialchars()) has been run.
1321 */
1322function format_to_edit( $content, $richedit = false ) {
1323	$content = apply_filters( 'format_to_edit', $content );
1324	if ( ! $richedit )
1325		$content = esc_textarea( $content );
1326	return $content;
1327}
1328
1329/**
1330 * Holder for the 'format_to_post' filter.
1331 *
1332 * @since 0.71
1333 *
1334 * @param string $content The text to pass through the filter.
1335 * @return string Text returned from the 'format_to_post' filter.
1336 */
1337function format_to_post($content) {
1338	$content = apply_filters('format_to_post', $content);
1339	return $content;
1340}
1341
1342/**
1343 * Add leading zeros when necessary.
1344 *
1345 * If you set the threshold to '4' and the number is '10', then you will get
1346 * back '0010'. If you set the threshold to '4' and the number is '5000', then you
1347 * will get back '5000'.
1348 *
1349 * Uses sprintf to append the amount of zeros based on the $threshold parameter
1350 * and the size of the number. If the number is large enough, then no zeros will
1351 * be appended.
1352 *
1353 * @since 0.71
1354 *
1355 * @param mixed $number Number to append zeros to if not greater than threshold.
1356 * @param int $threshold Digit places number needs to be to not have zeros added.
1357 * @return string Adds leading zeros to number if needed.
1358 */
1359function zeroise($number, $threshold) {
1360	return sprintf('%0'.$threshold.'s', $number);
1361}
1362
1363/**
1364 * Adds backslashes before letters and before a number at the start of a string.
1365 *
1366 * @since 0.71
1367 *
1368 * @param string $string Value to which backslashes will be added.
1369 * @return string String with backslashes inserted.
1370 */
1371function backslashit($string) {
1372	if ( isset( $string[0] ) && $string[0] >= '0' && $string[0] <= '9' )
1373		$string = '\\\\' . $string;
1374	return addcslashes( $string, 'A..Za..z' );
1375}
1376
1377/**
1378 * Appends a trailing slash.
1379 *
1380 * Will remove trailing slash if it exists already before adding a trailing
1381 * slash. This prevents double slashing a string or path.
1382 *
1383 * The primary use of this is for paths and thus should be used for paths. It is
1384 * not restricted to paths and offers no specific path support.
1385 *
1386 * @since 1.2.0
1387 * @uses untrailingslashit() Unslashes string if it was slashed already.
1388 *
1389 * @param string $string What to add the trailing slash to.
1390 * @return string String with trailing slash added.
1391 */
1392function trailingslashit($string) {
1393	return untrailingslashit($string) . '/';
1394}
1395
1396/**
1397 * Removes trailing slash if it exists.
1398 *
1399 * The primary use of this is for paths and thus should be used for paths. It is
1400 * not restricted to paths and offers no specific path support.
1401 *
1402 * @since 2.2.0
1403 *
1404 * @param string $string What to remove the trailing slash from.
1405 * @return string String without the trailing slash.
1406 */
1407function untrailingslashit($string) {
1408	return rtrim($string, '/');
1409}
1410
1411/**
1412 * Adds slashes to escape strings.
1413 *
1414 * Slashes will first be removed if magic_quotes_gpc is set, see {@link
1415 * http://www.php.net/magic_quotes} for more details.
1416 *
1417 * @since 0.71
1418 *
1419 * @param string $gpc The string returned from HTTP request data.
1420 * @return string Returns a string escaped with slashes.
1421 */
1422function addslashes_gpc($gpc) {
1423	if ( get_magic_quotes_gpc() )
1424		$gpc = stripslashes($gpc);
1425
1426	return wp_slash($gpc);
1427}
1428
1429/**
1430 * Navigates through an array and removes slashes from the values.
1431 *
1432 * If an array is passed, the array_map() function causes a callback to pass the
1433 * value back to the function. The slashes from this value will removed.
1434 *
1435 * @since 2.0.0
1436 *
1437 * @param mixed $value The value to be stripped.
1438 * @return mixed Stripped value.
1439 */
1440function stripslashes_deep($value) {
1441	if ( is_array($value) ) {
1442		$value = array_map('stripslashes_deep', $value);
1443	} elseif ( is_object($value) ) {
1444		$vars = get_object_vars( $value );
1445		foreach ($vars as $key=>$data) {
1446			$value->{$key} = stripslashes_deep( $data );
1447		}
1448	} elseif ( is_string( $value ) ) {
1449		$value = stripslashes($value);
1450	}
1451
1452	return $value;
1453}
1454
1455/**
1456 * Navigates through an array and encodes the values to be used in a URL.
1457 *
1458 *
1459 * @since 2.2.0
1460 *
1461 * @param array|string $value The array or string to be encoded.
1462 * @return array|string $value The encoded array (or string from the callback).
1463 */
1464function urlencode_deep($value) {
1465	$value = is_array($value) ? array_map('urlencode_deep', $value) : urlencode($value);
1466	return $value;
1467}
1468
1469/**
1470 * Navigates through an array and raw encodes the values to be used in a URL.
1471 *
1472 * @since 3.4.0
1473 *
1474 * @param array|string $value The array or string to be encoded.
1475 * @return array|string $value The encoded array (or string from the callback).
1476 */
1477function rawurlencode_deep( $value ) {
1478	return is_array( $value ) ? array_map( 'rawurlencode_deep', $value ) : rawurlencode( $value );
1479}
1480
1481/**
1482 * Converts email addresses characters to HTML entities to block spam bots.
1483 *
1484 * @since 0.71
1485 *
1486 * @param string $emailaddy Email address.
1487 * @param int $mailto Optional. Range from 0 to 1. Used for encoding.
1488 * @return string Converted email address.
1489 */
1490function antispambot($emailaddy, $mailto=0) {
1491	$emailNOSPAMaddy = '';
1492	srand ((float) microtime() * 1000000);
1493	for ($i = 0; $i < strlen($emailaddy); $i = $i + 1) {
1494		$j = floor(rand(0, 1+$mailto));
1495		if ($j==0) {
1496			$emailNOSPAMaddy .= '&#'.ord(substr($emailaddy,$i,1)).';';
1497		} elseif ($j==1) {
1498			$emailNOSPAMaddy .= substr($emailaddy,$i,1);
1499		} elseif ($j==2) {
1500			$emailNOSPAMaddy .= '%'.zeroise(dechex(ord(substr($emailaddy, $i, 1))), 2);
1501		}
1502	}
1503	$emailNOSPAMaddy = str_replace('@','&#64;',$emailNOSPAMaddy);
1504	return $emailNOSPAMaddy;
1505}
1506
1507/**
1508 * Callback to convert URI match to HTML A element.
1509 *
1510 * This function was backported from 2.5.0 to 2.3.2. Regex callback for {@link
1511 * make_clickable()}.
1512 *
1513 * @since 2.3.2
1514 * @access private
1515 *
1516 * @param array $matches Single Regex Match.
1517 * @return string HTML A element with URI address.
1518 */
1519function _make_url_clickable_cb($matches) {
1520	$url = $matches[2];
1521
1522	if ( ')' == $matches[3] && strpos( $url, '(' ) ) {
1523		// If the trailing character is a closing parethesis, and the URL has an opening parenthesis in it, add the closing parenthesis to the URL.
1524		// Then we can let the parenthesis balancer do its thing below.
1525		$url .= $matches[3];
1526		$suffix = '';
1527	} else {
1528		$suffix = $matches[3];
1529	}
1530
1531	// Include parentheses in the URL only if paired
1532	while ( substr_count( $url, '(' ) < substr_count( $url, ')' ) ) {
1533		$suffix = strrchr( $url, ')' ) . $suffix;
1534		$url = substr( $url, 0, strrpos( $url, ')' ) );
1535	}
1536
1537	$url = esc_url($url);
1538	if ( empty($url) )
1539		return $matches[0];
1540
1541	return $matches[1] . "<a href=\"$url\" rel=\"nofollow\">$url</a>" . $suffix;
1542}
1543
1544/**
1545 * Callback to convert URL match to HTML A element.
1546 *
1547 * This function was backported from 2.5.0 to 2.3.2. Regex callback for {@link
1548 * make_clickable()}.
1549 *
1550 * @since 2.3.2
1551 * @access private
1552 *
1553 * @param array $matches Single Regex Match.
1554 * @return string HTML A element with URL address.
1555 */
1556function _make_web_ftp_clickable_cb($matches) {
1557	$ret = '';
1558	$dest = $matches[2];
1559	$dest = 'http://' . $dest;
1560	$dest = esc_url($dest);
1561	if ( empty($dest) )
1562		return $matches[0];
1563
1564	// removed trailing [.,;:)] from URL
1565	if ( in_array( substr($dest, -1), array('.', ',', ';', ':', ')') ) === true ) {
1566		$ret = substr($dest, -1);
1567		$dest = substr($dest, 0, strlen($dest)-1);
1568	}
1569	return $matches[1] . "<a href=\"$dest\" rel=\"nofollow\">$dest</a>$ret";
1570}
1571
1572/**
1573 * Callback to convert email address match to HTML A element.
1574 *
1575 * This function was backported from 2.5.0 to 2.3.2. Regex callback for {@link
1576 * make_clickable()}.
1577 *
1578 * @since 2.3.2
1579 * @access private
1580 *
1581 * @param array $matches Single Regex Match.
1582 * @return string HTML A element with email address.
1583 */
1584function _make_email_clickable_cb($matches) {
1585	$email = $matches[2] . '@' . $matches[3];
1586	return $matches[1] . "<a href=\"mailto:$email\">$email</a>";
1587}
1588
1589/**
1590 * Convert plaintext URI to HTML links.
1591 *
1592 * Converts URI, www and ftp, and email addresses. Finishes by fixing links
1593 * within links.
1594 *
1595 * @since 0.71
1596 *
1597 * @param string $text Content to convert URIs.
1598 * @return string Content with converted URIs.
1599 */
1600function make_clickable( $text ) {
1601	$r = '';
1602	$textarr = preg_split( '/(<[^<>]+>)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE ); // split out HTML tags
1603	foreach ( $textarr as $piece ) {
1604		if ( empty( $piece ) || ( $piece[0] == '<' && ! preg_match('|^<\s*[\w]{1,20}+://|', $piece) ) ) {
1605			$r .= $piece;
1606			continue;
1607		}
1608
1609		// Long strings might contain expensive edge cases ...
1610		if ( 10000 < strlen( $piece ) ) {
1611			// ... break it up
1612			foreach ( _split_str_by_whitespace( $piece, 2100 ) as $chunk ) { // 2100: Extra room for scheme and leading and trailing paretheses
1613				if ( 2101 < strlen( $chunk ) ) {
1614					$r .= $chunk; // Too big, no whitespace: bail.
1615				} else {
1616					$r .= make_clickable( $chunk );
1617				}
1618			}
1619		} else {
1620			$ret = " $piece "; // Pad with whitespace to simplify the regexes
1621
1622			$url_clickable = '~
1623				([\\s(<.,;:!?])                                        # 1: Leading whitespace, or punctuation
1624				(                                                      # 2: URL
1625					[\\w]{1,20}+://                                # Scheme and hier-part prefix
1626					(?=\S{1,2000}\s)                               # Limit to URLs less than about 2000 characters long
1627					[\\w\\x80-\\xff#%\\~/@\\[\\]*(+=&$-]*+         # Non-punctuation URL character
1628					(?:                                            # Unroll the Loop: Only allow puctuation URL character if followed by a non-punctuation URL character
1629						[\'.,;:!?)]                            # Punctuation URL character
1630						[\\w\\x80-\\xff#%\\~/@\\[\\]*(+=&$-]++ # Non-punctuation URL character
1631					)*
1632				)
1633				(\)?)                                                  # 3: Trailing closing parenthesis (for parethesis balancing post processing)
1634			~xS'; // The regex is a non-anchored pattern and does not have a single fixed starting character.
1635			      // Tell PCRE to spend more time optimizing since, when used on a page load, it will probably be used several times.
1636
1637			$ret = preg_replace_callback( $url_clickable, '_make_url_clickable_cb', $ret );
1638
1639			$ret = preg_replace_callback( '#([\s>])((www|ftp)\.[\w\\x80-\\xff\#$%&~/.\-;:=,?@\[\]+]+)#is', '_make_web_ftp_clickable_cb', $ret );
1640			$ret = preg_replace_callback( '#([\s>])([.0-9a-z_+-]+)@(([0-9a-z-]+\.)+[0-9a-z]{2,})#i', '_make_email_clickable_cb', $ret );
1641
1642			$ret = substr( $ret, 1, -1 ); // Remove our whitespace padding.
1643			$r .= $ret;
1644		}
1645	}
1646
1647	// Cleanup of accidental links within links
1648	$r = preg_replace( '#(<a( [^>]+?>|>))<a [^>]+?>([^>]+?)</a></a>#i', "$1$3</a>", $r );
1649	return $r;
1650}
1651
1652/**
1653 * Breaks a string into chunks by splitting at whitespace characters.
1654 * The length of each returned chunk is as close to the specified length goal as possible,
1655 * with the caveat that each chunk includes its trailing delimiter.
1656 * Chunks longer than the goal are guaranteed to not have any inner whitespace.
1657 *
1658 * Joining the returned chunks with empty delimiters reconstructs the input string losslessly.
1659 *
1660 * Input string must have no null characters (or eventual transformations on output chunks must not care about null characters)
1661 *
1662 * <code>
1663 * _split_str_by_whitespace( "1234 67890 1234 67890a cd 1234   890 123456789 1234567890a    45678   1 3 5 7 90 ", 10 ) ==
1664 * array (
1665 *   0 => '1234 67890 ',  // 11 characters: Perfect split
1666 *   1 => '1234 ',        //  5 characters: '1234 67890a' was too long
1667 *   2 => '67890a cd ',   // 10 characters: '67890a cd 1234' was too long
1668 *   3 => '1234   890 ',  // 11 characters: Perfect split
1669 *   4 => '123456789 ',   // 10 characters: '123456789 1234567890a' was too long
1670 *   5 => '1234567890a ', // 12 characters: Too long, but no inner whitespace on which to split
1671 *   6 => '   45678   ',  // 11 characters: Perfect split
1672 *   7 => '1 3 5 7 9',    //  9 characters: End of $string
1673 * );
1674 * </code>
1675 *
1676 * @since 3.4.0
1677 * @access private
1678 *
1679 * @param string $string The string to split.
1680 * @param int $goal The desired chunk length.
1681 * @return array Numeric array of chunks.
1682 */
1683function _split_str_by_whitespace( $string, $goal ) {
1684	$chunks = array();
1685
1686	$string_nullspace = strtr( $string, "\r\n\t\v\f ", "\000\000\000\000\000\000" );
1687
1688	while ( $goal < strlen( $string_nullspace ) ) {
1689		$pos = strrpos( substr( $string_nullspace, 0, $goal + 1 ), "\000" );
1690
1691		if ( false === $pos ) {
1692			$pos = strpos( $string_nullspace, "\000", $goal + 1 );
1693			if ( false === $pos ) {
1694				break;
1695			}
1696		}
1697
1698		$chunks[] = substr( $string, 0, $pos + 1 );
1699		$string = substr( $string, $pos + 1 );
1700		$string_nullspace = substr( $string_nullspace, $pos + 1 );
1701	}
1702
1703	if ( $string ) {
1704		$chunks[] = $string;
1705	}
1706
1707	return $chunks;
1708}
1709
1710/**
1711 * Adds rel nofollow string to all HTML A elements in content.
1712 *
1713 * @since 1.5.0
1714 *
1715 * @param string $text Content that may contain HTML A elements.
1716 * @return string Converted content.
1717 */
1718function wp_rel_nofollow( $text ) {
1719	// This is a pre save filter, so text is already escaped.
1720	$text = stripslashes($text);
1721	$text = preg_replace_callback('|<a (.+?)>|i', 'wp_rel_nofollow_callback', $text);
1722	$text = wp_slash($text);
1723	return $text;
1724}
1725
1726/**
1727 * Callback to add rel=nofollow string to HTML A element.
1728 *
1729 * Will remove already existing rel="nofollow" and rel='nofollow' from the
1730 * string to prevent from invalidating (X)HTML.
1731 *
1732 * @since 2.3.0
1733 *
1734 * @param array $matches Single Match
1735 * @return string HTML A Element with rel nofollow.
1736 */
1737function wp_rel_nofollow_callback( $matches ) {
1738	$text = $matches[1];
1739	$text = str_replace(array(' rel="nofollow"', " rel='nofollow'"), '', $text);
1740	return "<a $text rel=\"nofollow\">";
1741}
1742
1743/**
1744 * Convert one smiley code to the icon graphic file equivalent.
1745 *
1746 * Looks up one smiley code in the $wpsmiliestrans global array and returns an
1747 * <img> string for that smiley.
1748 *
1749 * @global array $wpsmiliestrans
1750 * @since 2.8.0
1751 *
1752 * @param string $smiley Smiley code to convert to image.
1753 * @return string Image string for smiley.
1754 */
1755function translate_smiley($smiley) {
1756	global $wpsmiliestrans;
1757
1758	if (count($smiley) == 0) {
1759		return '';
1760	}
1761
1762	$smiley = trim(reset($smiley));
1763	$img = $wpsmiliestrans[$smiley];
1764	$smiley_masked = esc_attr($smiley);
1765
1766	$srcurl = apply_filters('smilies_src', includes_url("images/smilies/$img"), $img, site_url());
1767
1768	return " <img src='$srcurl' alt='$smiley_masked' class='wp-smiley' /> ";
1769}
1770
1771/**
1772 * Convert text equivalent of smilies to images.
1773 *
1774 * Will only convert smilies if the option 'use_smilies' is true and the global
1775 * used in the function isn't empty.
1776 *
1777 * @since 0.71
1778 * @uses $wp_smiliessearch
1779 *
1780 * @param string $text Content to convert smilies from text.
1781 * @return string Converted content with text smilies replaced with images.
1782 */
1783function convert_smilies($text) {
1784	global $wp_smiliessearch;
1785	$output = '';
1786	if ( get_option('use_smilies') && !empty($wp_smiliessearch) ) {
1787		// HTML loop taken from texturize function, could possible be consolidated
1788		$textarr = preg_split("/(<.*>)/U", $text, -1, PREG_SPLIT_DELIM_CAPTURE); // capture the tags as well as in between
1789		$stop = count($textarr);// loop stuff
1790		for ($i = 0; $i < $stop; $i++) {
1791			$content = $textarr[$i];
1792			if ((strlen($content) > 0) && ('<' != $content[0])) { // If it's not a tag
1793				$content = preg_replace_callback($wp_smiliessearch, 'translate_smiley', $content);
1794			}
1795			$output .= $content;
1796		}
1797	} else {
1798		// return default text.
1799		$output = $text;
1800	}
1801	return $output;
1802}
1803
1804/**
1805 * Verifies that an email is valid.
1806 *
1807 * Does not grok i18n domains. Not RFC compliant.
1808 *
1809 * @since 0.71
1810 *
1811 * @param string $email Email address to verify.
1812 * @param boolean $deprecated Deprecated.
1813 * @return string|bool Either false or the valid email address.
1814 */
1815function is_email( $email, $deprecated = false ) {
1816	if ( ! empty( $deprecated ) )
1817		_deprecated_argument( __FUNCTION__, '3.0' );
1818
1819	// Test for the minimum length the email can be
1820	if ( strlen( $email ) < 3 ) {
1821		return apply_filters( 'is_email', false, $email, 'email_too_short' );
1822	}
1823
1824	// Test for an @ character after the first position
1825	if ( strpos( $email, '@', 1 ) === false ) {
1826		return apply_filters( 'is_email', false, $email, 'email_no_at' );
1827	}
1828
1829	// Split out the local and domain parts
1830	list( $local, $domain ) = explode( '@', $email, 2 );
1831
1832	// LOCAL PART
1833	// Test for invalid characters
1834	if ( !preg_match( '/^[a-zA-Z0-9!#$%&\'*+\/=?^_`{|}~\.-]+$/', $local ) ) {
1835		return apply_filters( 'is_email', false, $email, 'local_invalid_chars' );
1836	}
1837
1838	// DOMAIN PART
1839	// Test for sequences of periods
1840	if ( preg_match( '/\.{2,}/', $domain ) ) {
1841		return apply_filters( 'is_email', false, $email, 'domain_period_sequence' );
1842	}
1843
1844	// Test for leading and trailing periods and whitespace
1845	if ( trim( $domain, " \t\n\r\0\x0B." ) !== $domain ) {
1846		return apply_filters( 'is_email', false, $email, 'domain_period_limits' );
1847	}
1848
1849	// Split the domain into subs
1850	$subs = explode( '.', $domain );
1851
1852	// Assume the domain will have at least two subs
1853	if ( 2 > count( $subs ) ) {
1854		return apply_filters( 'is_email', false, $email, 'domain_no_periods' );
1855	}
1856
1857	// Loop through each sub
1858	foreach ( $subs as $sub ) {
1859		// Test for leading and trailing hyphens and whitespace
1860		if ( trim( $sub, " \t\n\r\0\x0B-" ) !== $sub ) {
1861			return apply_filters( 'is_email', false, $email, 'sub_hyphen_limits' );
1862		}
1863
1864		// Test for invalid characters
1865		if ( !preg_match('/^[a-z0-9-]+$/i', $sub ) ) {
1866			return apply_filters( 'is_email', false, $email, 'sub_invalid_chars' );
1867		}
1868	}
1869
1870	// Congratulations your email made it!
1871	return apply_filters( 'is_email', $email, $email, null );
1872}
1873
1874/**
1875 * Convert to ASCII from email subjects.
1876 *
1877 * @since 1.2.0
1878 *
1879 * @param string $string Subject line
1880 * @return string Converted string to ASCII
1881 */
1882function wp_iso_descrambler($string) {
1883	/* this may only work with iso-8859-1, I'm afraid */
1884	if (!preg_match('#\=\?(.+)\?Q\?(.+)\?\=#i', $string, $matches)) {
1885		return $string;
1886	} else {
1887		$subject = str_replace('_', ' ', $matches[2]);
1888		$subject = preg_replace_callback('#\=([0-9a-f]{2})#i', '_wp_iso_convert', $subject);
1889		return $subject;
1890	}
1891}
1892
1893/**
1894 * Helper function to convert hex encoded chars to ASCII
1895 *
1896 * @since 3.1.0
1897 * @access private
1898 *
1899 * @param array $match The preg_replace_callback matches array
1900 * @return array Converted chars
1901 */
1902function _wp_iso_convert( $match ) {
1903	return chr( hexdec( strtolower( $match[1] ) ) );
1904}
1905
1906/**
1907 * Returns a date in the GMT equivalent.
1908 *
1909 * Requires and returns a date in the Y-m-d H:i:s format. If there is a
1910 * timezone_string available, the date is assumed to be in that timezone,
1911 * otherwise it simply subtracts the value of the 'gmt_offset' option. Return
1912 * format can be overridden using the $format parameter.
1913 *
1914 * @since 1.2.0
1915 *
1916 * @uses get_option() to retrieve the value of 'gmt_offset'.
1917 * @param string $string The date to be converted.
1918 * @param string $format The format string for the returned date (default is Y-m-d H:i:s)
1919 * @return string GMT version of the date provided.
1920 */
1921function get_gmt_from_date( $string, $format = 'Y-m-d H:i:s' ) {
1922	$tz = get_option( 'timezone_string' );
1923	if ( $tz ) {
1924		$datetime = date_create( $string, new DateTimeZone( $tz ) );
1925		if ( ! $datetime )
1926			return gmdate( $format, 0 );
1927		$datetime->setTimezone( new DateTimeZone( 'UTC' ) );
1928		$string_gmt = $datetime->format( $format );
1929	} else {
1930		if ( ! preg_match( '#([0-9]{1,4})-([0-9]{1,2})-([0-9]{1,2}) ([0-9]{1,2}):([0-9]{1,2}):([0-9]{1,2})#', $string, $matches ) )
1931			return gmdate( $format, 0 );
1932		$string_time = gmmktime( $matches[4], $matches[5], $matches[6], $matches[2], $matches[3], $matches[1] );
1933		$string_gmt = gmdate( $format, $string_time - get_option( 'gmt_offset' ) * HOUR_IN_SECONDS );
1934	}
1935	return $string_gmt;
1936}
1937
1938/**
1939 * Converts a GMT date into the correct format for the blog.
1940 *
1941 * Requires and returns a date in the Y-m-d H:i:s format. If there is a
1942 * timezone_string available, the returned date is in that timezone, otherwise
1943 * it simply adds the value of gmt_offset. Return format can be overridden
1944 * using the $format parameter
1945 *
1946 * @since 1.2.0
1947 *
1948 * @param string $string The date to be converted.
1949 * @param string $format The format string for the returned date (default is Y-m-d H:i:s)
1950 * @return string Formatted date relative to the timezone / GMT offset.
1951 */
1952function get_date_from_gmt( $string, $format = 'Y-m-d H:i:s' ) {
1953	$tz = get_option( 'timezone_string' );
1954	if ( $tz ) {
1955		$datetime = date_create( $string, new DateTimeZone( 'UTC' ) );
1956		if ( ! $datetime )
1957			return date( $format, 0 );
1958		$datetime->setTimezone( new DateTimeZone( $tz ) );
1959		$string_localtime = $datetime->format( $format );
1960	} else {
1961		if ( ! preg_match('#([0-9]{1,4})-([0-9]{1,2})-([0-9]{1,2}) ([0-9]{1,2}):([0-9]{1,2}):([0-9]{1,2})#', $string, $matches) )
1962			return date( $format, 0 );
1963		$string_time = gmmktime( $matches[4], $matches[5], $matches[6], $matches[2], $matches[3], $matches[1] );
1964		$string_localtime = gmdate( $format, $string_time + get_option( 'gmt_offset' ) * HOUR_IN_SECONDS );
1965	}
1966	return $string_localtime;
1967}
1968
1969/**
1970 * Computes an offset in seconds from an iso8601 timezone.
1971 *
1972 * @since 1.5.0
1973 *
1974 * @param string $timezone Either 'Z' for 0 offset or 'Âąhhmm'.
1975 * @return int|float The offset in seconds.
1976 */
1977function iso8601_timezone_to_offset($timezone) {
1978	// $timezone is either 'Z' or '[+|-]hhmm'
1979	if ($timezone == 'Z') {
1980		$offset = 0;
1981	} else {
1982		$sign    = (substr($timezone, 0, 1) == '+') ? 1 : -1;
1983		$hours   = intval(substr($timezone, 1, 2));
1984		$minutes = intval(substr($timezone, 3, 4)) / 60;
1985		$offset  = $sign * HOUR_IN_SECONDS * ($hours + $minutes);
1986	}
1987	return $offset;
1988}
1989
1990/**
1991 * Converts an iso8601 date to MySQL DateTime format used by post_date[_gmt].
1992 *
1993 * @since 1.5.0
1994 *
1995 * @param string $date_string Date and time in ISO 8601 format {@link http://en.wikipedia.org/wiki/ISO_8601}.
1996 * @param string $timezone Optional. If set to GMT returns the time minus gmt_offset. Default is 'user'.
1997 * @return string The date and time in MySQL DateTime format - Y-m-d H:i:s.
1998 */
1999function iso8601_to_datetime($date_string, $timezone = 'user') {
2000	$timezone = strtolower($timezone);
2001
2002	if ($timezone == 'gmt') {
2003
2004		preg_match('#([0-9]{4})([0-9]{2})([0-9]{2})T([0-9]{2}):([0-9]{2}):([0-9]{2})(Z|[\+|\-][0-9]{2,4}){0,1}#', $date_string, $date_bits);
2005
2006		if (!empty($date_bits[7])) { // we have a timezone, so let's compute an offset
2007			$offset = iso8601_timezone_to_offset($date_bits[7]);
2008		} else { // we don't have a timezone, so we assume user local timezone (not server's!)
2009			$offset = HOUR_IN_SECONDS * get_option('gmt_offset');
2010		}
2011
2012		$timestamp = gmmktime($date_bits[4], $date_bits[5], $date_bits[6], $date_bits[2], $date_bits[3], $date_bits[1]);
2013		$timestamp -= $offset;
2014
2015		return gmdate('Y-m-d H:i:s', $timestamp);
2016
2017	} else if ($timezone == 'user') {
2018		return preg_replace('#([0-9]{4})([0-9]{2})([0-9]{2})T([0-9]{2}):([0-9]{2}):([0-9]{2})(Z|[\+|\-][0-9]{2,4}){0,1}#', '$1-$2-$3 $4:$5:$6', $date_string);
2019	}
2020}
2021
2022/**
2023 * Adds a element attributes to open links in new windows.
2024 *
2025 * Comment text in popup windows should be filtered through this. Right now it's
2026 * a moderately dumb function, ideally it would detect whether a target or rel
2027 * attribute was already there and adjust its actions accordingly.
2028 *
2029 * @since 0.71
2030 *
2031 * @param string $text Content to replace links to open in a new window.
2032 * @return string Content that has filtered links.
2033 */
2034function popuplinks($text) {
2035	$text = preg_replace('/<a (.+?)>/i', "<a $1 target='_blank' rel='external'>", $text);
2036	return $text;
2037}
2038
2039/**
2040 * Strips out all characters that are not allowable in an email.
2041 *
2042 * @since 1.5.0
2043 *
2044 * @param string $email Email address to filter.
2045 * @return string Filtered email address.
2046 */
2047function sanitize_email( $email ) {
2048	// Test for the minimum length the email can be
2049	if ( strlen( $email ) < 3 ) {
2050		return apply_filters( 'sanitize_email', '', $email, 'email_too_short' );
2051	}
2052
2053	// Test for an @ character after the first position
2054	if ( strpos( $email, '@', 1 ) === false ) {
2055		return apply_filters( 'sanitize_email', '', $email, 'email_no_at' );
2056	}
2057
2058	// Split out the local and domain parts
2059	list( $local, $domain ) = explode( '@', $email, 2 );
2060
2061	// LOCAL PART
2062	// Test for invalid characters
2063	$local = preg_replace( '/[^a-zA-Z0-9!#$%&\'*+\/=?^_`{|}~\.-]/', '', $local );
2064	if ( '' === $local ) {
2065		return apply_filters( 'sanitize_email', '', $email, 'local_invalid_chars' );
2066	}
2067
2068	// DOMAIN PART
2069	// Test for sequences of periods
2070	$domain = preg_replace( '/\.{2,}/', '', $domain );
2071	if ( '' === $domain ) {
2072		return apply_filters( 'sanitize_email', '', $email, 'domain_period_sequence' );
2073	}
2074
2075	// Test for leading and trailing periods and whitespace
2076	$domain = trim( $domain, " \t\n\r\0\x0B." );
2077	if ( '' === $domain ) {
2078		return apply_filters( 'sanitize_email', '', $email, 'domain_period_limits' );
2079	}
2080
2081	// Split the domain into subs
2082	$subs = explode( '.', $domain );
2083
2084	// Assume the domain will have at least two subs
2085	if ( 2 > count( $subs ) ) {
2086		return apply_filters( 'sanitize_email', '', $email, 'domain_no_periods' );
2087	}
2088
2089	// Create an array that will contain valid subs
2090	$new_subs = array();
2091
2092	// Loop through each sub
2093	foreach ( $subs as $sub ) {
2094		// Test for leading and trailing hyphens
2095		$sub = trim( $sub, " \t\n\r\0\x0B-" );
2096
2097		// Test for invalid characters
2098		$sub = preg_replace( '/[^a-z0-9-]+/i', '', $sub );
2099
2100		// If there's anything left, add it to the valid subs
2101		if ( '' !== $sub ) {
2102			$new_subs[] = $sub;
2103		}
2104	}
2105
2106	// If there aren't 2 or more valid subs
2107	if ( 2 > count( $new_subs ) ) {
2108		return apply_filters( 'sanitize_email', '', $email, 'domain_no_valid_subs' );
2109	}
2110
2111	// Join valid subs into the new domain
2112	$domain = join( '.', $new_subs );
2113
2114	// Put the email back together
2115	$email = $local . '@' . $domain;
2116
2117	// Congratulations your email made it!
2118	return apply_filters( 'sanitize_email', $email, $email, null );
2119}
2120
2121/**
2122 * Determines the difference between two timestamps.
2123 *
2124 * The difference is returned in a human readable format such as "1 hour",
2125 * "5 mins", "2 days".
2126 *
2127 * @since 1.5.0
2128 *
2129 * @param int $from Unix timestamp from which the difference begins.
2130 * @param int $to Optional. Unix timestamp to end the time difference. Default becomes time() if not set.
2131 * @return string Human readable time difference.
2132 */
2133function human_time_diff( $from, $to = '' ) {
2134	if ( empty( $to ) )
2135		$to = time();
2136	$diff = (int) abs( $to - $from );
2137	if ( $diff <= HOUR_IN_SECONDS ) {
2138		$mins = round( $diff / MINUTE_IN_SECONDS );
2139		if ( $mins <= 1 ) {
2140			$mins = 1;
2141		}
2142		/* translators: min=minute */
2143		$since = sprintf( _n( '%s min', '%s mins', $mins ), $mins );
2144	} elseif ( ( $diff <= DAY_IN_SECONDS ) && ( $diff > HOUR_IN_SECONDS ) ) {
2145		$hours = round( $diff / HOUR_IN_SECONDS );
2146		if ( $hours <= 1 ) {
2147			$hours = 1;
2148		}
2149		$since = sprintf( _n( '%s hour', '%s hours', $hours ), $hours );
2150	} elseif ( $diff >= DAY_IN_SECONDS ) {
2151		$days = round( $diff / DAY_IN_SECONDS );
2152		if ( $days <= 1 ) {
2153			$days = 1;
2154		}
2155		$since = sprintf( _n( '%s day', '%s days', $days ), $days );
2156	}
2157	return $since;
2158}
2159
2160/**
2161 * Generates an excerpt from the content, if needed.
2162 *
2163 * The excerpt word amount will be 55 words and if the amount is greater than
2164 * that, then the string ' [&hellip;]' will be appended to the excerpt. If the string
2165 * is less than 55 words, then the content will be returned as is.
2166 *
2167 * The 55 word limit can be modified by plugins/themes using the excerpt_length filter
2168 * The ' [&hellip;]' string can be modified by plugins/themes using the excerpt_more filter
2169 *
2170 * @since 1.5.0
2171 *
2172 * @param string $text Optional. The excerpt. If set to empty, an excerpt is generated.
2173 * @return string The excerpt.
2174 */
2175function wp_trim_excerpt($text = '') {
2176	$raw_excerpt = $text;
2177	if ( '' == $text ) {
2178		$text = get_the_content('');
2179
2180		$text = strip_shortcodes( $text );
2181
2182		$text = apply_filters('the_content', $text);
2183		$text = str_replace(']]>', ']]&gt;', $text);
2184		$excerpt_length = apply_filters('excerpt_length', 55);
2185		$excerpt_more = apply_filters('excerpt_more', ' ' . '[&hellip;]');
2186		$text = wp_trim_words( $text, $excerpt_length, $excerpt_more );
2187	}
2188	return apply_filters('wp_trim_excerpt', $text, $raw_excerpt);
2189}
2190
2191/**
2192 * Trims text to a certain number of words.
2193 *
2194 * This function is localized. For languages that count 'words' by the individual
2195 * character (such as East Asian languages), the $num_words argument will apply
2196 * to the number of individual characters.
2197 *
2198 * @since 3.3.0
2199 *
2200 * @param string $text Text to trim.
2201 * @param int $num_words Number of words. Default 55.
2202 * @param string $more Optional. What to append if $text needs to be trimmed. Default '&hellip;'.
2203 * @return string Trimmed text.
2204 */
2205function wp_trim_words( $text, $num_words = 55, $more = null ) {
2206	if ( null === $more )
2207		$more = __( '&hellip;' );
2208	$original_text = $text;
2209	$text = wp_strip_all_tags( $text );
2210	/* translators: If your word count is based on single characters (East Asian characters),
2211	   enter 'characters'. Otherwise, enter 'words'. Do not translate into your own language. */
2212	if ( 'characters' == _x( 'words', 'word count: words or characters?' ) && preg_match( '/^utf\-?8$/i', get_option( 'blog_charset' ) ) ) {
2213		$text = trim( preg_replace( "/[\n\r\t ]+/", ' ', $text ), ' ' );
2214		preg_match_all( '/./u', $text, $words_array );
2215		$words_array = array_slice( $words_array[0], 0, $num_words + 1 );
2216		$sep = '';
2217	} else {
2218		$words_array = preg_split( "/[\n\r\t ]+/", $text, $num_words + 1, PREG_SPLIT_NO_EMPTY );
2219		$sep = ' ';
2220	}
2221	if ( count( $words_array ) > $num_words ) {
2222		array_pop( $words_array );
2223		$text = implode( $sep, $words_array );
2224		$text = $text . $more;
2225	} else {
2226		$text = implode( $sep, $words_array );
2227	}
2228	return apply_filters( 'wp_trim_words', $text, $num_words, $more, $original_text );
2229}
2230
2231/**
2232 * Converts named entities into numbered entities.
2233 *
2234 * @since 1.5.1
2235 *
2236 * @param string $text The text within which entities will be converted.
2237 * @return string Text with converted entities.
2238 */
2239function ent2ncr($text) {
2240
2241	// Allow a plugin to short-circuit and override the mappings.
2242	$filtered = apply_filters( 'pre_ent2ncr', null, $text );
2243	if( null !== $filtered )
2244		return $filtered;
2245
2246	$to_ncr = array(
2247		'&quot;' => '&#34;',
2248		'&amp;' => '&#38;',
2249		'&lt;' => '&#60;',
2250		'&gt;' => '&#62;',
2251		'|' => '&#124;',
2252		'&nbsp;' => '&#160;',
2253		'&iexcl;' => '&#161;',
2254		'&cent;' => '&#162;',
2255		'&pound;' => '&#163;',
2256		'&curren;' => '&#164;',
2257		'&yen;' => '&#165;',
2258		'&brvbar;' => '&#166;',
2259		'&brkbar;' => '&#166;',
2260		'&sect;' => '&#167;',
2261		'&uml;' => '&#168;',
2262		'&die;' => '&#168;',
2263		'&copy;' => '&#169;',
2264		'&ordf;' => '&#170;',
2265		'&laquo;' => '&#171;',
2266		'&not;' => '&#172;',
2267		'&shy;' => '&#173;',
2268		'&reg;' => '&#174;',
2269		'&macr;' => '&#175;',
2270		'&hibar;' => '&#175;',
2271		'&deg;' => '&#176;',
2272		'&plusmn;' => '&#177;',
2273		'&sup2;' => '&#178;',
2274		'&sup3;' => '&#179;',
2275		'&acute;' => '&#180;',
2276		'&micro;' => '&#181;',
2277		'&para;' => '&#182;',
2278		'&middot;' => '&#183;',
2279		'&cedil;' => '&#184;',
2280		'&sup1;' => '&#185;',
2281		'&ordm;' => '&#186;',
2282		'&raquo;' => '&#187;',
2283		'&frac14;' => '&#188;',
2284		'&frac12;' => '&#189;',
2285		'&frac34;' => '&#190;',
2286		'&iquest;' => '&#191;',
2287		'&Agrave;' => '&#192;',
2288		'&Aacute;' => '&#193;',
2289		'&Acirc;' => '&#194;',
2290		'&Atilde;' => '&#195;',
2291		'&Auml;' => '&#196;',
2292		'&Aring;' => '&#197;',
2293		'&AElig;' => '&#198;',
2294		'&Ccedil;' => '&#199;',
2295		'&Egrave;' => '&#200;',
2296		'&Eacute;' => '&#201;',
2297		'&Ecirc;' => '&#202;',
2298		'&Euml;' => '&#203;',
2299		'&Igrave;' => '&#204;',
2300		'&Iacute;' => '&#205;',
2301		'&Icirc;' => '&#206;',
2302		'&Iuml;' => '&#207;',
2303		'&ETH;' => '&#208;',
2304		'&Ntilde;' => '&#209;',
2305		'&Ograve;' => '&#210;',
2306		'&Oacute;' => '&#211;',
2307		'&Ocirc;' => '&#212;',
2308		'&Otilde;' => '&#213;',
2309		'&Ouml;' => '&#214;',
2310		'&times;' => '&#215;',
2311		'&Oslash;' => '&#216;',
2312		'&Ugrave;' => '&#217;',
2313		'&Uacute;' => '&#218;',
2314		'&Ucirc;' => '&#219;',
2315		'&Uuml;' => '&#220;',
2316		'&Yacute;' => '&#221;',
2317		'&THORN;' => '&#222;',
2318		'&szlig;' => '&#223;',
2319		'&agrave;' => '&#224;',
2320		'&aacute;' => '&#225;',
2321		'&acirc;' => '&#226;',
2322		'&atilde;' => '&#227;',
2323		'&auml;' => '&#228;',
2324		'&aring;' => '&#229;',
2325		'&aelig;' => '&#230;',
2326		'&ccedil;' => '&#231;',
2327		'&egrave;' => '&#232;',
2328		'&eacute;' => '&#233;',
2329		'&ecirc;' => '&#234;',
2330		'&euml;' => '&#235;',
2331		'&igrave;' => '&#236;',
2332		'&iacute;' => '&#237;',
2333		'&icirc;' => '&#238;',
2334		'&iuml;' => '&#239;',
2335		'&eth;' => '&#240;',
2336		'&ntilde;' => '&#241;',
2337		'&ograve;' => '&#242;',
2338		'&oacute;' => '&#243;',
2339		'&ocirc;' => '&#244;',
2340		'&otilde;' => '&#245;',
2341		'&ouml;' => '&#246;',
2342		'&divide;' => '&#247;',
2343		'&oslash;' => '&#248;',
2344		'&ugrave;' => '&#249;',
2345		'&uacute;' => '&#250;',
2346		'&ucirc;' => '&#251;',
2347		'&uuml;' => '&#252;',
2348		'&yacute;' => '&#253;',
2349		'&thorn;' => '&#254;',
2350		'&yuml;' => '&#255;',
2351		'&OElig;' => '&#338;',
2352		'&oelig;' => '&#339;',
2353		'&Scaron;' => '&#352;',
2354		'&scaron;' => '&#353;',
2355		'&Yuml;' => '&#376;',
2356		'&fnof;' => '&#402;',
2357		'&circ;' => '&#710;',
2358		'&tilde;' => '&#732;',
2359		'&Alpha;' => '&#913;',
2360		'&Beta;' => '&#914;',
2361		'&Gamma;' => '&#915;',
2362		'&Delta;' => '&#916;',
2363		'&Epsilon;' => '&#917;',
2364		'&Zeta;' => '&#918;',
2365		'&Eta;' => '&#919;',
2366		'&Theta;' => '&#920;',
2367		'&Iota;' => '&#921;',
2368		'&Kappa;' => '&#922;',
2369		'&Lambda;' => '&#923;',
2370		'&Mu;' => '&#924;',
2371		'&Nu;' => '&#925;',
2372		'&Xi;' => '&#926;',
2373		'&Omicron;' => '&#927;',
2374		'&Pi;' => '&#928;',
2375		'&Rho;' => '&#929;',
2376		'&Sigma;' => '&#931;',
2377		'&Tau;' => '&#932;',
2378		'&Upsilon;' => '&#933;',
2379		'&Phi;' => '&#934;',
2380		'&Chi;' => '&#935;',
2381		'&Psi;' => '&#936;',
2382		'&Omega;' => '&#937;',
2383		'&alpha;' => '&#945;',
2384		'&beta;' => '&#946;',
2385		'&gamma;' => '&#947;',
2386		'&delta;' => '&#948;',
2387		'&epsilon;' => '&#949;',
2388		'&zeta;' => '&#950;',
2389		'&eta;' => '&#951;',
2390		'&theta;' => '&#952;',
2391		'&iota;' => '&#953;',
2392		'&kappa;' => '&#954;',
2393		'&lambda;' => '&#955;',
2394		'&mu;' => '&#956;',
2395		'&nu;' => '&#957;',
2396		'&xi;' => '&#958;',
2397		'&omicron;' => '&#959;',
2398		'&pi;' => '&#960;',
2399		'&rho;' => '&#961;',
2400		'&sigmaf;' => '&#962;',
2401		'&sigma;' => '&#963;',
2402		'&tau;' => '&#964;',
2403		'&upsilon;' => '&#965;',
2404		'&phi;' => '&#966;',
2405		'&chi;' => '&#967;',
2406		'&psi;' => '&#968;',
2407		'&omega;' => '&#969;',
2408		'&thetasym;' => '&#977;',
2409		'&upsih;' => '&#978;',
2410		'&piv;' => '&#982;',
2411		'&ensp;' => '&#8194;',
2412		'&emsp;' => '&#8195;',
2413		'&thinsp;' => '&#8201;',
2414		'&zwnj;' => '&#8204;',
2415		'&zwj;' => '&#8205;',
2416		'&lrm;' => '&#8206;',
2417		'&rlm;' => '&#8207;',
2418		'&ndash;' => '&#8211;',
2419		'&mdash;' => '&#8212;',
2420		'&lsquo;' => '&#8216;',
2421		'&rsquo;' => '&#8217;',
2422		'&sbquo;' => '&#8218;',
2423		'&ldquo;' => '&#8220;',
2424		'&rdquo;' => '&#8221;',
2425		'&bdquo;' => '&#8222;',
2426		'&dagger;' => '&#8224;',
2427		'&Dagger;' => '&#8225;',
2428		'&bull;' => '&#8226;',
2429		'&hellip;' => '&#8230;',
2430		'&permil;' => '&#8240;',
2431		'&prime;' => '&#8242;',
2432		'&Prime;' => '&#8243;',
2433		'&lsaquo;' => '&#8249;',
2434		'&rsaquo;' => '&#8250;',
2435		'&oline;' => '&#8254;',
2436		'&frasl;' => '&#8260;',
2437		'&euro;' => '&#8364;',
2438		'&image;' => '&#8465;',
2439		'&weierp;' => '&#8472;',
2440		'&real;' => '&#8476;',
2441		'&trade;' => '&#8482;',
2442		'&alefsym;' => '&#8501;',
2443		'&crarr;' => '&#8629;',
2444		'&lArr;' => '&#8656;',
2445		'&uArr;' => '&#8657;',
2446		'&rArr;' => '&#8658;',
2447		'&dArr;' => '&#8659;',
2448		'&hArr;' => '&#8660;',
2449		'&forall;' => '&#8704;',
2450		'&part;' => '&#8706;',
2451		'&exist;' => '&#8707;',
2452		'&empty;' => '&#8709;',
2453		'&nabla;' => '&#8711;',
2454		'&isin;' => '&#8712;',
2455		'&notin;' => '&#8713;',
2456		'&ni;' => '&#8715;',
2457		'&prod;' => '&#8719;',
2458		'&sum;' => '&#8721;',
2459		'&minus;' => '&#8722;',
2460		'&lowast;' => '&#8727;',
2461		'&radic;' => '&#8730;',
2462		'&prop;' => '&#8733;',
2463		'&infin;' => '&#8734;',
2464		'&ang;' => '&#8736;',
2465		'&and;' => '&#8743;',
2466		'&or;' => '&#8744;',
2467		'&cap;' => '&#8745;',
2468		'&cup;' => '&#8746;',
2469		'&int;' => '&#8747;',
2470		'&there4;' => '&#8756;',
2471		'&sim;' => '&#8764;',
2472		'&cong;' => '&#8773;',
2473		'&asymp;' => '&#8776;',
2474		'&ne;' => '&#8800;',
2475		'&equiv;' => '&#8801;',
2476		'&le;' => '&#8804;',
2477		'&ge;' => '&#8805;',
2478		'&sub;' => '&#8834;',
2479		'&sup;' => '&#8835;',
2480		'&nsub;' => '&#8836;',
2481		'&sube;' => '&#8838;',
2482		'&supe;' => '&#8839;',
2483		'&oplus;' => '&#8853;',
2484		'&otimes;' => '&#8855;',
2485		'&perp;' => '&#8869;',
2486		'&sdot;' => '&#8901;',
2487		'&lceil;' => '&#8968;',
2488		'&rceil;' => '&#8969;',
2489		'&lfloor;' => '&#8970;',
2490		'&rfloor;' => '&#8971;',
2491		'&lang;' => '&#9001;',
2492		'&rang;' => '&#9002;',
2493		'&larr;' => '&#8592;',
2494		'&uarr;' => '&#8593;',
2495		'&rarr;' => '&#8594;',
2496		'&darr;' => '&#8595;',
2497		'&harr;' => '&#8596;',
2498		'&loz;' => '&#9674;',
2499		'&spades;' => '&#9824;',
2500		'&clubs;' => '&#9827;',
2501		'&hearts;' => '&#9829;',
2502		'&diams;' => '&#9830;'
2503	);
2504
2505	return str_replace( array_keys($to_ncr), array_values($to_ncr), $text );
2506}
2507
2508/**
2509 * Formats text for the rich text editor.
2510 *
2511 * The filter 'richedit_pre' is applied here. If $text is empty the filter will
2512 * be applied to an empty string.
2513 *
2514 * @since 2.0.0
2515 *
2516 * @param string $text The text to be formatted.
2517 * @return string The formatted text after filter is applied.
2518 */
2519function wp_richedit_pre($text) {
2520	// Filtering a blank results in an annoying <br />\n
2521	if ( empty($text) ) return apply_filters('richedit_pre', '');
2522
2523	$output = convert_chars($text);
2524	$output = wpautop($output);
2525	$output = htmlspecialchars($output, ENT_NOQUOTES, get_option( 'blog_charset' ) );
2526
2527	return apply_filters('richedit_pre', $output);
2528}
2529
2530/**
2531 * Formats text for the HTML editor.
2532 *
2533 * Unless $output is empty it will pass through htmlspecialchars before the
2534 * 'htmledit_pre' filter is applied.
2535 *
2536 * @since 2.5.0
2537 *
2538 * @param string $output The text to be formatted.
2539 * @return string Formatted text after filter applied.
2540 */
2541function wp_htmledit_pre($output) {
2542	if ( !empty($output) )
2543		$output = htmlspecialchars($output, ENT_NOQUOTES, get_option( 'blog_charset' ) ); // convert only < > &
2544
2545	return apply_filters('htmledit_pre', $output);
2546}
2547
2548/**
2549 * Perform a deep string replace operation to ensure the values in $search are no longer present
2550 *
2551 * Repeats the replacement operation until it no longer replaces anything so as to remove "nested" values
2552 * e.g. $subject = '%0%0%0DDD', $search ='%0D', $result ='' rather than the '%0%0DD' that
2553 * str_replace would return
2554 *
2555 * @since 2.8.1
2556 * @access private
2557 *
2558 * @param string|array $search
2559 * @param string $subject
2560 * @return string The processed string
2561 */
2562function _deep_replace( $search, $subject ) {
2563	$found = true;
2564	$subject = (string) $subject;
2565	while ( $found ) {
2566		$found = false;
2567		foreach ( (array) $search as $val ) {
2568			while ( strpos( $subject, $val ) !== false ) {
2569				$found = true;
2570				$subject = str_replace( $val, '', $subject );
2571			}
2572		}
2573	}
2574
2575	return $subject;
2576}
2577
2578/**
2579 * Escapes data for use in a MySQL query
2580 *
2581 * This is just a handy shortcut for $wpdb->escape(), for completeness' sake
2582 *
2583 * @since 2.8.0
2584 * @param string $sql Unescaped SQL data
2585 * @return string The cleaned $sql
2586 */
2587function esc_sql( $sql ) {
2588	global $wpdb;
2589	return $wpdb->escape( $sql );
2590}
2591
2592/**
2593 * Checks and cleans a URL.
2594 *
2595 * A number of characters are removed from the URL. If the URL is for displaying
2596 * (the default behaviour) ampersands are also replaced. The 'clean_url' filter
2597 * is applied to the returned cleaned URL.
2598 *
2599 * @since 2.8.0
2600 * @uses wp_kses_bad_protocol() To only permit protocols in the URL set
2601 *		via $protocols or the common ones set in the function.
2602 *
2603 * @param string $url The URL to be cleaned.
2604 * @param array $protocols Optional. An array of acceptable protocols.
2605 *		Defaults to 'http', 'https', 'ftp', 'ftps', 'mailto', 'news', 'irc', 'gopher', 'nntp', 'feed', 'telnet', 'mms', 'rtsp', 'svn' if not set.
2606 * @param string $_context Private. Use esc_url_raw() for database usage.
2607 * @return string The cleaned $url after the 'clean_url' filter is applied.
2608 */
2609function esc_url( $url, $protocols = null, $_context = 'display' ) {
2610	$original_url = $url;
2611
2612	if ( '' == $url )
2613		return $url;
2614	$url = preg_replace('|[^a-z0-9-~+_.?#=!&;,/:%@$\|*\'()\\x80-\\xff]|i', '', $url);
2615	$strip = array('%0d', '%0a', '%0D', '%0A');
2616	$url = _deep_replace($strip, $url);
2617	$url = str_replace(';//', '://', $url);
2618	/* If the URL doesn't appear to contain a scheme, we
2619	 * presume it needs http:// appended (unless a relative
2620	 * link starting with /, # or ? or a php file).
2621	 */
2622	if ( strpos($url, ':') === false && ! in_array( $url[0], array( '/', '#', '?' ) ) &&
2623		! preg_match('/^[a-z0-9-]+?\.php/i', $url) )
2624		$url = 'http://' . $url;
2625
2626	// Replace ampersands and single quotes only when displaying.
2627	if ( 'display' == $_context ) {
2628		$url = wp_kses_normalize_entities( $url );
2629		$url = str_replace( '&amp;', '&#038;', $url );
2630		$url = str_replace( "'", '&#039;', $url );
2631	}
2632
2633	if ( ! is_array( $protocols ) )
2634		$protocols = wp_allowed_protocols();
2635	$good_protocol_url = wp_kses_bad_protocol( $url, $protocols );
2636	if ( strtolower( $good_protocol_url ) != strtolower( $url ) )
2637		return '';
2638
2639	return apply_filters('clean_url', $good_protocol_url, $original_url, $_context);
2640}
2641
2642/**
2643 * Performs esc_url() for database usage.
2644 *
2645 * @since 2.8.0
2646 * @uses esc_url()
2647 *
2648 * @param string $url The URL to be cleaned.
2649 * @param array $protocols An array of acceptable protocols.
2650 * @return string The cleaned URL.
2651 */
2652function esc_url_raw( $url, $protocols = null ) {
2653	return esc_url( $url, $protocols, 'db' );
2654}
2655
2656/**
2657 * Convert entities, while preserving already-encoded entities.
2658 *
2659 * @link http://www.php.net/htmlentities Borrowed from the PHP Manual user notes.
2660 *
2661 * @since 1.2.2
2662 *
2663 * @param string $myHTML The text to be converted.
2664 * @return string Converted text.
2665 */
2666function htmlentities2($myHTML) {
2667	$translation_table = get_html_translation_table( HTML_ENTITIES, ENT_QUOTES );
2668	$translation_table[chr(38)] = '&';
2669	return preg_replace( "/&(?![A-Za-z]{0,4}\w{2,3};|#[0-9]{2,3};)/", "&amp;", strtr($myHTML, $translation_table) );
2670}
2671
2672/**
2673 * Escape single quotes, htmlspecialchar " < > &, and fix line endings.
2674 *
2675 * Escapes text strings for echoing in JS. It is intended to be used for inline JS
2676 * (in a tag attribute, for example onclick="..."). Note that the strings have to
2677 * be in single quotes. The filter 'js_escape' is also applied here.
2678 *
2679 * @since 2.8.0
2680 *
2681 * @param string $text The text to be escaped.
2682 * @return string Escaped text.
2683 */
2684function esc_js( $text ) {
2685	$safe_text = wp_check_invalid_utf8( $text );
2686	$safe_text = _wp_specialchars( $safe_text, ENT_COMPAT );
2687	$safe_text = preg_replace( '/&#(x)?0*(?(1)27|39);?/i', "'", stripslashes( $safe_text ) );
2688	$safe_text = str_replace( "\r", '', $safe_text );
2689	$safe_text = str_replace( "\n", '\\n', addslashes( $safe_text ) );
2690	return apply_filters( 'js_escape', $safe_text, $text );
2691}
2692
2693/**
2694 * Escaping for HTML blocks.
2695 *
2696 * @since 2.8.0
2697 *
2698 * @param string $text
2699 * @return string
2700 */
2701function esc_html( $text ) {
2702	$safe_text = wp_check_invalid_utf8( $text );
2703	$safe_text = _wp_specialchars( $safe_text, ENT_QUOTES );
2704	return apply_filters( 'esc_html', $safe_text, $text );
2705}
2706
2707/**
2708 * Escaping for HTML attributes.
2709 *
2710 * @since 2.8.0
2711 *
2712 * @param string $text
2713 * @return string
2714 */
2715function esc_attr( $text ) {
2716	$safe_text = wp_check_invalid_utf8( $text );
2717	$safe_text = _wp_specialchars( $safe_text, ENT_QUOTES );
2718	return apply_filters( 'attribute_escape', $safe_text, $text );
2719}
2720
2721/**
2722 * Escaping for textarea values.
2723 *
2724 * @since 3.1.0
2725 *
2726 * @param string $text
2727 * @return string
2728 */
2729function esc_textarea( $text ) {
2730	$safe_text = htmlspecialchars( $text, ENT_QUOTES, get_option( 'blog_charset' ) );
2731	return apply_filters( 'esc_textarea', $safe_text, $text );
2732}
2733
2734/**
2735 * Escape an HTML tag name.
2736 *
2737 * @since 2.5.0
2738 *
2739 * @param string $tag_name
2740 * @return string
2741 */
2742function tag_escape($tag_name) {
2743	$safe_tag = strtolower( preg_replace('/[^a-zA-Z0-9_:]/', '', $tag_name) );
2744	return apply_filters('tag_escape', $safe_tag, $tag_name);
2745}
2746
2747/**
2748 * Escapes text for SQL LIKE special characters % and _.
2749 *
2750 * @since 2.5.0
2751 *
2752 * @param string $text The text to be escaped.
2753 * @return string text, safe for inclusion in LIKE query.
2754 */
2755function like_escape($text) {
2756	return str_replace(array("%", "_"), array("\\%", "\\_"), $text);
2757}
2758
2759/**
2760 * Convert full URL paths to absolute paths.
2761 *
2762 * Removes the http or https protocols and the domain. Keeps the path '/' at the
2763 * beginning, so it isn't a true relative link, but from the web root base.
2764 *
2765 * @since 2.1.0
2766 *
2767 * @param string $link Full URL path.
2768 * @return string Absolute path.
2769 */
2770function wp_make_link_relative( $link ) {
2771	return preg_replace( '|https?://[^/]+(/.*)|i', '$1', $link );
2772}
2773
2774/**
2775 * Sanitises various option values based on the nature of the option.
2776 *
2777 * This is basically a switch statement which will pass $value through a number
2778 * of functions depending on the $option.
2779 *
2780 * @since 2.0.5
2781 *
2782 * @param string $option The name of the option.
2783 * @param string $value The unsanitised value.
2784 * @return string Sanitized value.
2785 */
2786function sanitize_option($option, $value) {
2787
2788	switch ( $option ) {
2789		case 'admin_email' :
2790		case 'new_admin_email' :
2791			$value = sanitize_email( $value );
2792			if ( ! is_email( $value ) ) {
2793				$value = get_option( $option ); // Resets option to stored value in the case of failed sanitization
2794				if ( function_exists( 'add_settings_error' ) )
2795					add_settings_error( $option, 'invalid_admin_email', __( 'The email address entered did not appear to be a valid email address. Please enter a valid email address.' ) );
2796			}
2797			break;
2798
2799		case 'thumbnail_size_w':
2800		case 'thumbnail_size_h':
2801		case 'medium_size_w':
2802		case 'medium_size_h':
2803		case 'large_size_w':
2804		case 'large_size_h':
2805		case 'mailserver_port':
2806		case 'comment_max_links':
2807		case 'page_on_front':
2808		case 'page_for_posts':
2809		case 'rss_excerpt_length':
2810		case 'default_category':
2811		case 'default_email_category':
2812		case 'default_link_category':
2813		case 'close_comments_days_old':
2814		case 'comments_per_page':
2815		case 'thread_comments_depth':
2816		case 'users_can_register':
2817		case 'start_of_week':
2818			$value = absint( $value );
2819			break;
2820
2821		case 'posts_per_page':
2822		case 'posts_per_rss':
2823			$value = (int) $value;
2824			if ( empty($value) )
2825				$value = 1;
2826			if ( $value < -1 )
2827				$value = abs($value);
2828			break;
2829
2830		case 'default_ping_status':
2831		case 'default_comment_status':
2832			// Options that if not there have 0 value but need to be something like "closed"
2833			if ( $value == '0' || $value == '')
2834				$value = 'closed';
2835			break;
2836
2837		case 'blogdescription':
2838		case 'blogname':
2839			$value = wp_kses_post( $value );
2840			$value = esc_html( $value );
2841			break;
2842
2843		case 'blog_charset':
2844			$value = preg_replace('/[^a-zA-Z0-9_-]/', '', $value); // strips slashes
2845			break;
2846
2847		case 'blog_public':
2848			// This is the value if the settings checkbox is not checked on POST. Don't rely on this.
2849			if ( null === $value )
2850				$value = 1;
2851			else
2852				$value = intval( $value );
2853			break;
2854
2855		case 'date_format':
2856		case 'time_format':
2857		case 'mailserver_url':
2858		case 'mailserver_login':
2859		case 'mailserver_pass':
2860		case 'upload_path':
2861			$value = strip_tags( $value );
2862			$value = wp_kses_data( $value );
2863			break;
2864
2865		case 'ping_sites':
2866			$value = explode( "\n", $value );
2867			$value = array_filter( array_map( 'trim', $value ) );
2868			$value = array_filter( array_map( 'esc_url_raw', $value ) );
2869			$value = implode( "\n", $value );
2870			break;
2871
2872		case 'gmt_offset':
2873			$value = preg_replace('/[^0-9:.-]/', '', $value); // strips slashes
2874			break;
2875
2876		case 'siteurl':
2877			if ( (bool)preg_match( '#http(s?)://(.+)#i', $value) ) {
2878				$value = esc_url_raw($value);
2879			} else {
2880				$value = get_option( $option ); // Resets option to stored value in the case of failed sanitization
2881				if ( function_exists('add_settings_error') )
2882					add_settings_error('siteurl', 'invalid_siteurl', __('The WordPress address you entered did not appear to be a valid URL. Please enter a valid URL.'));
2883			}
2884			break;
2885
2886		case 'home':
2887			if ( (bool)preg_match( '#http(s?)://(.+)#i', $value) ) {
2888				$value = esc_url_raw($value);
2889			} else {
2890				$value = get_option( $option ); // Resets option to stored value in the case of failed sanitization
2891				if ( function_exists('add_settings_error') )
2892					add_settings_error('home', 'invalid_home', __('The Site address you entered did not appear to be a valid URL. Please enter a valid URL.'));
2893			}
2894			break;
2895
2896		case 'WPLANG':
2897			$allowed = get_available_languages();
2898			if ( ! in_array( $value, $allowed ) && ! empty( $value ) )
2899				$value = get_option( $option );
2900			break;
2901
2902		case 'illegal_names':
2903			if ( ! is_array( $value ) )
2904				$value = explode( ' ', $value );
2905
2906			$value = array_values( array_filter( array_map( 'trim', $value ) ) );
2907
2908			if ( ! $value )
2909				$value = '';
2910			break;
2911
2912		case 'limited_email_domains':
2913		case 'banned_email_domains':
2914			if ( ! is_array( $value ) )
2915				$value = explode( "\n", $value );
2916
2917			$domains = array_values( array_filter( array_map( 'trim', $value ) ) );
2918			$value = array();
2919
2920			foreach ( $domains as $domain ) {
2921				if ( ! preg_match( '/(--|\.\.)/', $domain ) && preg_match( '|^([a-zA-Z0-9-\.])+$|', $domain ) )
2922					$value[] = $domain;
2923			}
2924			if ( ! $value )
2925				$value = '';
2926			break;
2927
2928		case 'timezone_string':
2929			$allowed_zones = timezone_identifiers_list();
2930			if ( ! in_array( $value, $allowed_zones ) && ! empty( $value ) ) {
2931				$value = get_option( $option ); // Resets option to stored value in the case of failed sanitization
2932				if ( function_exists('add_settings_error') )
2933					add_settings_error('timezone_string', 'invalid_timezone_string', __('The timezone you have entered is not valid. Please select a valid timezone.') );
2934			}
2935			break;
2936
2937		case 'permalink_structure':
2938		case 'category_base':
2939		case 'tag_base':
2940			$value = esc_url_raw( $value );
2941			$value = str_replace( 'http://', '', $value );
2942			break;
2943	}
2944
2945	$value = apply_filters("sanitize_option_{$option}", $value, $option);
2946
2947	return $value;
2948}
2949
2950/**
2951 * Parses a string into variables to be stored in an array.
2952 *
2953 * Uses {@link http://www.php.net/parse_str parse_str()} and stripslashes if
2954 * {@link http://www.php.net/magic_quotes magic_quotes_gpc} is on.
2955 *
2956 * @since 2.2.1
2957 * @uses apply_filters() for the 'wp_parse_str' filter.
2958 *
2959 * @param string $string The string to be parsed.
2960 * @param array $array Variables will be stored in this array.
2961 */
2962function wp_parse_str( $string, &$array ) {
2963	parse_str( $string, $array );
2964	if ( get_magic_quotes_gpc() )
2965		$array = stripslashes_deep( $array );
2966	$array = apply_filters( 'wp_parse_str', $array );
2967}
2968
2969/**
2970 * Convert lone less than signs.
2971 *
2972 * KSES already converts lone greater than signs.
2973 *
2974 * @uses wp_pre_kses_less_than_callback in the callback function.
2975 * @since 2.3.0
2976 *
2977 * @param string $text Text to be converted.
2978 * @return string Converted text.
2979 */
2980function wp_pre_kses_less_than( $text ) {
2981	return preg_replace_callback('%<[^>]*?((?=<)|>|$)%', 'wp_pre_kses_less_than_callback', $text);
2982}
2983
2984/**
2985 * Callback function used by preg_replace.
2986 *
2987 * @uses esc_html to format the $matches text.
2988 * @since 2.3.0
2989 *
2990 * @param array $matches Populated by matches to preg_replace.
2991 * @return string The text returned after esc_html if needed.
2992 */
2993function wp_pre_kses_less_than_callback( $matches ) {
2994	if ( false === strpos($matches[0], '>') )
2995		return esc_html($matches[0]);
2996	return $matches[0];
2997}
2998
2999/**
3000 * WordPress implementation of PHP sprintf() with filters.
3001 *
3002 * @since 2.5.0
3003 * @link http://www.php.net/sprintf
3004 *
3005 * @param string $pattern The string which formatted args are inserted.
3006 * @param mixed $args,... Arguments to be formatted into the $pattern string.
3007 * @return string The formatted string.
3008 */
3009function wp_sprintf( $pattern ) {
3010	$args = func_get_args();
3011	$len = strlen($pattern);
3012	$start = 0;
3013	$result = '';
3014	$arg_index = 0;
3015	while ( $len > $start ) {
3016		// Last character: append and break
3017		if ( strlen($pattern) - 1 == $start ) {
3018			$result .= substr($pattern, -1);
3019			break;
3020		}
3021
3022		// Literal %: append and continue
3023		if ( substr($pattern, $start, 2) == '%%' ) {
3024			$start += 2;
3025			$result .= '%';
3026			continue;
3027		}
3028
3029		// Get fragment before next %
3030		$end = strpos($pattern, '%', $start + 1);
3031		if ( false === $end )
3032			$end = $len;
3033		$fragment = substr($pattern, $start, $end - $start);
3034
3035		// Fragment has a specifier
3036		if ( $pattern[$start] == '%' ) {
3037			// Find numbered arguments or take the next one in order
3038			if ( preg_match('/^%(\d+)\$/', $fragment, $matches) ) {
3039				$arg = isset($args[$matches[1]]) ? $args[$matches[1]] : '';
3040				$fragment = str_replace("%{$matches[1]}$", '%', $fragment);
3041			} else {
3042				++$arg_index;
3043				$arg = isset($args[$arg_index]) ? $args[$arg_index] : '';
3044			}
3045
3046			// Apply filters OR sprintf
3047			$_fragment = apply_filters( 'wp_sprintf', $fragment, $arg );
3048			if ( $_fragment != $fragment )
3049				$fragment = $_fragment;
3050			else
3051				$fragment = sprintf($fragment, strval($arg) );
3052		}
3053
3054		// Append to result and move to next fragment
3055		$result .= $fragment;
3056		$start = $end;
3057	}
3058	return $result;
3059}
3060
3061/**
3062 * Localize list items before the rest of the content.
3063 *
3064 * The '%l' must be at the first characters can then contain the rest of the
3065 * content. The list items will have ', ', ', and', and ' and ' added depending
3066 * on the amount of list items in the $args parameter.
3067 *
3068 * @since 2.5.0
3069 *
3070 * @param string $pattern Content containing '%l' at the beginning.
3071 * @param array $args List items to prepend to the content and replace '%l'.
3072 * @return string Localized list items and rest of the content.
3073 */
3074function wp_sprintf_l($pattern, $args) {
3075	// Not a match
3076	if ( substr($pattern, 0, 2) != '%l' )
3077		return $pattern;
3078
3079	// Nothing to work with
3080	if ( empty($args) )
3081		return '';
3082
3083	// Translate and filter the delimiter set (avoid ampersands and entities here)
3084	$l = apply_filters('wp_sprintf_l', array(
3085		/* translators: used between list items, there is a space after the comma */
3086		'between'          => __(', '),
3087		/* translators: used between list items, there is a space after the and */
3088		'between_last_two' => __(', and '),
3089		/* translators: used between only two list items, there is a space after the and */
3090		'between_only_two' => __(' and '),
3091		));
3092
3093	$args = (array) $args;
3094	$result = array_shift($args);
3095	if ( count($args) == 1 )
3096		$result .= $l['between_only_two'] . array_shift($args);
3097	// Loop when more than two args
3098	$i = count($args);
3099	while ( $i ) {
3100		$arg = array_shift($args);
3101		$i--;
3102		if ( 0 == $i )
3103			$result .= $l['between_last_two'] . $arg;
3104		else
3105			$result .= $l['between'] . $arg;
3106	}
3107	return $result . substr($pattern, 2);
3108}
3109
3110/**
3111 * Safely extracts not more than the first $count characters from html string.
3112 *
3113 * UTF-8, tags and entities safe prefix extraction. Entities inside will *NOT*
3114 * be counted as one character. For example &amp; will be counted as 4, &lt; as
3115 * 3, etc.
3116 *
3117 * @since 2.5.0
3118 *
3119 * @param string $str String to get the excerpt from.
3120 * @param integer $count Maximum number of characters to take.
3121 * @param string $more Optional. What to append if $str needs to be trimmed. Defaults to empty string.
3122 * @return string The excerpt.
3123 */
3124function wp_html_excerpt( $str, $count, $more = null ) {
3125	if ( null === $more )
3126		$more = '';
3127	$str = wp_strip_all_tags( $str, true );
3128	$excerpt = mb_substr( $str, 0, $count );
3129	// remove part of an entity at the end
3130	$excerpt = preg_replace( '/&[^;\s]{0,6}$/', '', $excerpt );
3131	if ( $str != $excerpt )
3132		$excerpt = trim( $excerpt ) . $more;
3133	return $excerpt;
3134}
3135
3136/**
3137 * Add a Base url to relative links in passed content.
3138 *
3139 * By default it supports the 'src' and 'href' attributes. However this can be
3140 * changed via the 3rd param.
3141 *
3142 * @since 2.7.0
3143 *
3144 * @param string $content String to search for links in.
3145 * @param string $base The base URL to prefix to links.
3146 * @param array $attrs The attributes which should be processed.
3147 * @return string The processed content.
3148 */
3149function links_add_base_url( $content, $base, $attrs = array('src', 'href') ) {
3150	global $_links_add_base;
3151	$_links_add_base = $base;
3152	$attrs = implode('|', (array)$attrs);
3153	return preg_replace_callback( "!($attrs)=(['\"])(.+?)\\2!i", '_links_add_base', $content );
3154}
3155
3156/**
3157 * Callback to add a base url to relative links in passed content.
3158 *
3159 * @since 2.7.0
3160 * @access private
3161 *
3162 * @param string $m The matched link.
3163 * @return string The processed link.
3164 */
3165function _links_add_base($m) {
3166	global $_links_add_base;
3167	//1 = attribute name  2 = quotation mark  3 = URL
3168	return $m[1] . '=' . $m[2] .
3169		( preg_match( '#^(\w{1,20}):#', $m[3], $protocol ) && in_array( $protocol[1], wp_allowed_protocols() ) ?
3170			$m[3] :
3171			path_join( $_links_add_base, $m[3] ) )
3172		. $m[2];
3173}
3174
3175/**
3176 * Adds a Target attribute to all links in passed content.
3177 *
3178 * This function by default only applies to <a> tags, however this can be
3179 * modified by the 3rd param.
3180 *
3181 * <b>NOTE:</b> Any current target attributed will be stripped and replaced.
3182 *
3183 * @since 2.7.0
3184 *
3185 * @param string $content String to search for links in.
3186 * @param string $target The Target to add to the links.
3187 * @param array $tags An array of tags to apply to.
3188 * @return string The processed content.
3189 */
3190function links_add_target( $content, $target = '_blank', $tags = array('a') ) {
3191	global $_links_add_target;
3192	$_links_add_target = $target;
3193	$tags = implode('|', (array)$tags);
3194	return preg_replace_callback( "!<($tags)(.+?)>!i", '_links_add_target', $content );
3195}
3196
3197/**
3198 * Callback to add a target attribute to all links in passed content.
3199 *
3200 * @since 2.7.0
3201 * @access private
3202 *
3203 * @param string $m The matched link.
3204 * @return string The processed link.
3205 */
3206function _links_add_target( $m ) {
3207	global $_links_add_target;
3208	$tag = $m[1];
3209	$link = preg_replace('|(target=([\'"])(.*?)\2)|i', '', $m[2]);
3210	return '<' . $tag . $link . ' target="' . esc_attr( $_links_add_target ) . '">';
3211}
3212
3213/**
3214 * Normalize EOL characters and strip duplicate whitespace.
3215 *
3216 * @since 2.7.0
3217 *
3218 * @param string $str The string to normalize.
3219 * @return string The normalized string.
3220 */
3221function normalize_whitespace( $str ) {
3222	$str  = trim( $str );
3223	$str  = str_replace( "\r", "\n", $str );
3224	$str  = preg_replace( array( '/\n+/', '/[ \t]+/' ), array( "\n", ' ' ), $str );
3225	return $str;
3226}
3227
3228/**
3229 * Properly strip all HTML tags including script and style
3230 *
3231 * @since 2.9.0
3232 *
3233 * @param string $string String containing HTML tags
3234 * @param bool $remove_breaks optional Whether to remove left over line breaks and white space chars
3235 * @return string The processed string.
3236 */
3237function wp_strip_all_tags($string, $remove_breaks = false) {
3238	$string = preg_replace( '@<(script|style)[^>]*?>.*?</\\1>@si', '', $string );
3239	$string = strip_tags($string);
3240
3241	if ( $remove_breaks )
3242		$string = preg_replace('/[\r\n\t ]+/', ' ', $string);
3243
3244	return trim( $string );
3245}
3246
3247/**
3248 * Sanitize a string from user input or from the db
3249 *
3250 * check for invalid UTF-8,
3251 * Convert single < characters to entity,
3252 * strip all tags,
3253 * remove line breaks, tabs and extra white space,
3254 * strip octets.
3255 *
3256 * @since 2.9.0
3257 *
3258 * @param string $str
3259 * @return string
3260 */
3261function sanitize_text_field($str) {
3262	$filtered = wp_check_invalid_utf8( $str );
3263
3264	if ( strpos($filtered, '<') !== false ) {
3265		$filtered = wp_pre_kses_less_than( $filtered );
3266		// This will strip extra whitespace for us.
3267		$filtered = wp_strip_all_tags( $filtered, true );
3268	} else {
3269		$filtered = trim( preg_replace('/[\r\n\t ]+/', ' ', $filtered) );
3270	}
3271
3272	$found = false;
3273	while ( preg_match('/%[a-f0-9]{2}/i', $filtered, $match) ) {
3274		$filtered = str_replace($match[0], '', $filtered);
3275		$found = true;
3276	}
3277
3278	if ( $found ) {
3279		// Strip out the whitespace that may now exist after removing the octets.
3280		$filtered = trim( preg_replace('/ +/', ' ', $filtered) );
3281	}
3282
3283	return apply_filters('sanitize_text_field', $filtered, $str);
3284}
3285
3286/**
3287 * i18n friendly version of basename()
3288 *
3289 * @since 3.1.0
3290 *
3291 * @param string $path A path.
3292 * @param string $suffix If the filename ends in suffix this will also be cut off.
3293 * @return string
3294 */
3295function wp_basename( $path, $suffix = '' ) {
3296	return urldecode( basename( str_replace( array( '%2F', '%5C' ), '/', urlencode( $path ) ), $suffix ) );
3297}
3298
3299/**
3300 * Forever eliminate "Wordpress" from the planet (or at least the little bit we can influence).
3301 *
3302 * Violating our coding standards for a good function name.
3303 *
3304 * @since 3.0.0
3305 */
3306function capital_P_dangit( $text ) {
3307	// Simple replacement for titles
3308	if ( 'the_title' === current_filter() )
3309		return str_replace( 'Wordpress', 'WordPress', $text );
3310	// Still here? Use the more judicious replacement
3311	static $dblq = false;
3312	if ( false === $dblq )
3313		$dblq = _x( '&#8220;', 'opening curly double quote' );
3314	return str_replace(
3315		array( ' Wordpress', '&#8216;Wordpress', $dblq . 'Wordpress', '>Wordpress', '(Wordpress' ),
3316		array( ' WordPress', '&#8216;WordPress', $dblq . 'WordPress', '>WordPress', '(WordPress' ),
3317	$text );
3318
3319}
3320
3321/**
3322 * Sanitize a mime type
3323 *
3324 * @since 3.1.3
3325 *
3326 * @param string $mime_type Mime type
3327 * @return string Sanitized mime type
3328 */
3329function sanitize_mime_type( $mime_type ) {
3330	$sani_mime_type = preg_replace( '/[^-+*.a-zA-Z0-9\/]/', '', $mime_type );
3331	return apply_filters( 'sanitize_mime_type', $sani_mime_type, $mime_type );
3332}
3333
3334/**
3335 * Sanitize space or carriage return separated URLs that are used to send trackbacks.
3336 *
3337 * @since 3.4.0
3338 *
3339 * @param string $to_ping Space or carriage return separated URLs
3340 * @return string URLs starting with the http or https protocol, separated by a carriage return.
3341 */
3342function sanitize_trackback_urls( $to_ping ) {
3343	$urls_to_ping = preg_split( '/[\r\n\t ]/', trim( $to_ping ), -1, PREG_SPLIT_NO_EMPTY );
3344	foreach ( $urls_to_ping as $k => $url ) {
3345		if ( !preg_match( '#^https?://.#i', $url ) )
3346			unset( $urls_to_ping[$k] );
3347	}
3348	$urls_to_ping = array_map( 'esc_url_raw', $urls_to_ping );
3349	$urls_to_ping = implode( "\n", $urls_to_ping );
3350	return apply_filters( 'sanitize_trackback_urls', $urls_to_ping, $to_ping );
3351}
3352
3353/**
3354 * Add slashes to a string or array of strings.
3355 *
3356 * This should be used when preparing data for core API that expects slashed data.
3357 * This should not be used to escape data going directly into an SQL query.
3358 *
3359 * @since 3.6.0
3360 *
3361 * @param string|array $value String or array of strings to slash.
3362 * @return string|array Slashed $value
3363 */
3364function wp_slash( $value ) {
3365	if ( is_array( $value ) ) {
3366		foreach ( $value as $k => $v ) {
3367			if ( is_array( $v ) ) {
3368				$value[$k] = wp_slash( $v );
3369			} else {
3370				$value[$k] = addslashes( $v );
3371			}
3372		}
3373	} else {
3374		$value = addslashes( $value );
3375	}
3376
3377	return $value;
3378}
3379
3380/**
3381 * Remove slashes from a string or array of strings.
3382 *
3383 * This should be used to remove slashes from data passed to core API that
3384 * expects data to be unslashed.
3385 *
3386 * @since 3.6.0
3387 *
3388 * @param string|array $value String or array of strings to unslash.
3389 * @return string|array Unslashed $value
3390 */
3391function wp_unslash( $value ) {
3392	return stripslashes_deep( $value );
3393}