3 +--------------------------------------------------------------------+
4 | CiviCRM version 4.6 |
5 +--------------------------------------------------------------------+
6 | Copyright CiviCRM LLC (c) 2004-2014 |
7 +--------------------------------------------------------------------+
8 | This file is a part of CiviCRM. |
10 | CiviCRM is free software; you can copy, modify, and distribute it |
11 | under the terms of the GNU Affero General Public License |
12 | Version 3, 19 November 2007 and the CiviCRM Licensing Exception. |
14 | CiviCRM is distributed in the hope that it will be useful, but |
15 | WITHOUT ANY WARRANTY; without even the implied warranty of |
16 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. |
17 | See the GNU Affero General Public License for more details. |
19 | You should have received a copy of the GNU Affero General Public |
20 | License and the CiviCRM Licensing Exception along |
21 | with this program; if not, contact CiviCRM LLC |
22 | at info[AT]civicrm[DOT]org. If you have questions about the |
23 | GNU Affero General Public License or the licensing of CiviCRM, |
24 | see the CiviCRM license FAQ at http://civicrm.org/licensing |
25 +--------------------------------------------------------------------+
31 * @copyright CiviCRM LLC (c) 2004-2014
36 require_once 'HTML/QuickForm/Rule/Email.php';
39 * This class contains string functions
42 class CRM_Utils_String
{
43 const COMMA
= ",", SEMICOLON
= ";", SPACE
= " ", TAB
= "\t", LINEFEED
= "\n", CARRIAGELINE
= "\r\n", LINECARRIAGE
= "\n\r", CARRIAGERETURN
= "\r";
46 * List of all letters and numbers
48 const ALPHANUMERIC
= 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890';
51 * Convert a display name into a potential variable
52 * name that we could use in forms/code
54 * @param name Name of the string
56 * @param int $maxLength
58 * @return string An equivalent variable name
61 * @return string (or null)
64 public static function titleToVar($title, $maxLength = 31) {
65 $variable = self
::munge($title, '_', $maxLength);
67 if (CRM_Utils_Rule
::title($variable, $maxLength)) {
71 // if longer than the maxLength lets just return a substr of the
72 // md5 to prevent errors downstream
73 return substr(md5($title), 0, $maxLength);
77 * Given a string, replace all non alpha numeric characters and
78 * spaces with the replacement character
81 * The name to be worked on.
83 * The character to use for non-valid chars.
85 * Length of valid variables.
88 * @return string returns the manipulated string
91 public static function munge($name, $char = '_', $len = 63) {
92 // replace all white space and non-alpha numeric with $char
93 // we only use the ascii character set since mysql does not create table names / field names otherwise
95 $name = preg_replace('/[^a-zA-Z0-9]+/', $char, trim($name));
98 // lets keep variable names short
99 return substr($name, 0, $len);
107 * Convert possibly underscore separated words to camel case with special handling for 'UF'
109 * membership_payment returns MembershipPayment
110 * @param string $string
112 * @return string string
114 public static function convertStringToCamel($string) {
115 $fragments = explode('_', $string);
116 foreach ($fragments as & $fragment) {
117 $fragment = ucfirst($fragment);
119 // Special case: UFGroup, UFJoin, UFMatch, UFField
120 if ($fragments[0] === 'Uf') {
121 $fragments[0] = 'UF';
123 return implode('', $fragments);
128 * Takes a variable name and munges it randomly into another variable name
130 * @param string $name
131 * Initial Variable Name.
133 * Length of valid variables.
135 * @return string Randomized Variable Name
138 public static function rename($name, $len = 4) {
139 $rand = substr(uniqid(), 0, $len);
140 return substr_replace($name, $rand, -$len, $len);
144 * Takes a string and returns the last tuple of the string.
145 * useful while converting file names to class names etc
147 * @param string $string
149 * @param \char|string $char $char the character used to demarcate the componets
152 * @return string the last component
155 public static function getClassName($string, $char = '_') {
157 if (!is_array($string)) {
158 $names = explode($char, $string);
160 if (!empty($names)) {
161 return array_pop($names);
166 * Appends a name to a string and seperated by delimiter.
167 * does the right thing for an empty string
170 * The string to be appended to.
171 * @param string $delim
172 * The delimiter to use.
174 * The string (or array of strings) to append.
179 public static function append(&$str, $delim, $name) {
184 if (is_array($name)) {
185 foreach ($name as $n) {
202 $str .= $delim . $name;
208 * Determine if the string is composed only of ascii characters
213 * Attempt utf8 match on failure (default yes).
215 * @return boolean true if string is ascii
218 public static function isAscii($str, $utf8 = TRUE) {
219 if (!function_exists('mb_detect_encoding')) {
220 // eliminate all white space from the string
221 $str = preg_replace('/\s+/', '', $str);
222 // FIXME: This is a pretty brutal hack to make utf8 and 8859-1 work.
224 /* match low- or high-ascii characters */
225 if (preg_match('/[\x00-\x20]|[\x7F-\xFF]/', $str)) {
226 // || // low ascii characters
227 // high ascii characters
228 // preg_match( '/[\x7F-\xFF]/', $str ) ) {
230 /* if we did match, try for utf-8, or iso8859-1 */
232 return self
::isUtf8($str);
241 $order = array('ASCII');
245 $enc = mb_detect_encoding($str, $order, TRUE);
246 return ($enc == 'ASCII' ||
$enc == 'UTF-8');
251 * Determine the string replacements for redaction
252 * on the basis of the regular expressions
256 * @param array $regexRules
257 * Regular expression to be matched w/ replacements.
259 * @return array $match array of strings w/ corresponding redacted outputs
262 public static function regex($str, $regexRules) {
263 //redact the regular expressions
264 if (!empty($regexRules) && isset($str)) {
265 static $matches, $totalMatches, $match = array();
266 foreach ($regexRules as $pattern => $replacement) {
267 preg_match_all($pattern, $str, $matches);
268 if (!empty($matches[0])) {
269 if (empty($totalMatches)) {
270 $totalMatches = $matches[0];
273 $totalMatches = array_merge($totalMatches, $matches[0]);
275 $match = array_flip($totalMatches);
280 if (!empty($match)) {
281 foreach ($match as $matchKey => & $dontCare) {
282 foreach ($regexRules as $pattern => $replacement) {
283 if (preg_match($pattern, $matchKey)) {
284 $dontCare = $replacement . substr(md5($matchKey), 0, 5);
291 return CRM_Core_DAO
::$_nullArray;
296 * @param $stringRules
300 public static function redaction($str, $stringRules) {
302 if (!empty($stringRules)) {
303 foreach ($stringRules as $match => $replace) {
304 $str = str_ireplace($match, $replace, $str);
308 //return the redacted output
313 * Determine if a string is composed only of utf8 characters
321 public static function isUtf8($str) {
322 if (!function_exists(mb_detect_encoding
)) {
323 // eliminate all white space from the string
324 $str = preg_replace('/\s+/', '', $str);
326 /* pattern stolen from the php.net function documentation for
328 * comment by JF Sebastian, 30-Mar-2005
331 return preg_match('/^([\x00-\x7f]|[\xc2-\xdf][\x80-\xbf]|\xe0[\xa0-\xbf][\x80-\xbf]|[\xe1-\xec][\x80-\xbf]{2}|\xed[\x80-\x9f][\x80-\xbf]|[\xee-\xef][\x80-\xbf]{2}|f0[\x90-\xbf][\x80-\xbf]{2}|[\xf1-\xf3][\x80-\xbf]{3}|\xf4[\x80-\x8f][\x80-\xbf]{2})*$/', $str);
333 // iconv('ISO-8859-1', 'UTF-8', $str);
336 $enc = mb_detect_encoding($str, array('UTF-8'), TRUE);
337 return ($enc !== FALSE);
342 * Determine if two href's are equivalent (fuzzy match)
344 * @param string $url1
345 * The first url to be matched.
346 * @param string $url2
347 * The second url to be matched against.
349 * @return boolean true if the urls match, else false
352 public static function match($url1, $url2) {
353 $url1 = strtolower($url1);
354 $url2 = strtolower($url2);
356 $url1Str = parse_url($url1);
357 $url2Str = parse_url($url2);
359 if ($url1Str['path'] == $url2Str['path'] &&
360 self
::extractURLVarValue(CRM_Utils_Array
::value('query', $url1Str)) == self
::extractURLVarValue(CRM_Utils_Array
::value('query', $url2Str))
368 * Extract variable values
371 * This is basically url.
373 * @return mix $v returns civicrm url (eg: civicrm/contact/search/...)
376 public static function extractURLVarValue($query) {
377 $config = CRM_Core_Config
::singleton();
378 $urlVar = $config->userFrameworkURLVar
;
380 $params = explode('&', $query);
381 foreach ($params as $p) {
382 if (strpos($p, '=')) {
383 list($k, $v) = explode('=', $p);
393 * Translate a true/false/yes/no string to a 0 or 1 value
396 * The string to be translated.
401 public static function strtobool($str) {
402 if (!is_scalar($str)) {
406 if (preg_match('/^(y(es)?|t(rue)?|1)$/i', $str)) {
413 * Returns string '1' for a true/yes/1 string, and '0' for no/false/0 else returns false
416 * The string to be translated.
421 public static function strtoboolstr($str) {
422 if (!is_scalar($str)) {
426 if (preg_match('/^(y(es)?|t(rue)?|1)$/i', $str)) {
429 elseif (preg_match('/^(n(o)?|f(alse)?|0)$/i', $str)) {
438 * Convert a HTML string into a text one using html2text
440 * @param string $html
441 * The string to be converted.
443 * @return string the converted string
446 public static function htmlToText($html) {
447 require_once 'packages/html2text/rcube_html2text.php';
448 $token_html = preg_replace('!\{([a-z_.]+)\}!i', 'token:{$1}', $html);
449 $converter = new rcube_html2text($token_html);
450 $token_text = $converter->get_text();
451 $text = preg_replace('!token\:\{([a-z_.]+)\}!i', '{$1}', $token_text);
457 * @param array $params
459 public static function extractName($string, &$params) {
460 $name = trim($string);
466 $name = str_replace('"', '', $name);
467 $name = str_replace('\'', '', $name);
469 // check for comma in name
470 if (strpos($name, ',') !== FALSE) {
472 // name has a comma - assume lname, fname [mname]
473 $names = explode(',', $name);
474 if (count($names) > 1) {
475 $params['last_name'] = trim($names[0]);
477 // check for space delim
478 $fnames = explode(' ', trim($names[1]));
479 if (count($fnames) > 1) {
480 $params['first_name'] = trim($fnames[0]);
481 $params['middle_name'] = trim($fnames[1]);
484 $params['first_name'] = trim($fnames[0]);
488 $params['first_name'] = trim($names[0]);
492 // name has no comma - assume fname [mname] fname
493 $names = explode(' ', $name);
494 if (count($names) == 1) {
495 $params['first_name'] = $names[0];
497 elseif (count($names) == 2) {
498 $params['first_name'] = $names[0];
499 $params['last_name'] = $names[1];
502 $params['first_name'] = $names[0];
503 $params['middle_name'] = $names[1];
504 $params['last_name'] = $names[2];
514 public static function &makeArray($string) {
515 $string = trim($string);
517 $values = explode("\n", $string);
519 foreach ($values as $value) {
520 list($n, $v) = CRM_Utils_System
::explode('=', $value, 2);
522 $result[trim($n)] = trim($v);
529 * Given an ezComponents-parsed representation of
530 * a text with alternatives return only the first one
532 * @param string $full
533 * All alternatives as a long string (or some other text).
535 * @return string only the first alternative found (or the text without alternatives)
537 public static function stripAlternatives($full) {
539 preg_match('/-ALTERNATIVE ITEM 0-(.*?)-ALTERNATIVE ITEM 1-.*-ALTERNATIVE END-/s', $full, $matches);
541 if (isset($matches[1]) &&
542 trim(strip_tags($matches[1])) != ''
552 * Strip leading, trailing, double spaces from string
553 * used for postal/greeting/addressee
555 * @param string $string
556 * Input string to be cleaned.
558 * @return string the cleaned string
561 public static function stripSpaces($string) {
562 return (empty($string)) ?
$string : preg_replace("/\s{2,}/", " ", trim($string));
566 * This function is used to clean the URL 'path' variable that we use
567 * to construct CiviCRM urls by removing characters from the path variable
569 * @param string $string
570 * The input string to be sanitized.
571 * @param array $search
572 * The characters to be sanitized.
573 * @param string $replace
574 * The character to replace it with.
576 * @return string the sanitized string
579 static function stripPathChars($string,
583 static $_searchChars = NULL;
584 static $_replaceChar = NULL;
586 if (empty($string)) {
590 if ($_searchChars == NULL) {
591 $_searchChars = array(
592 '&', ';', ',', '=', '$',
595 ' ', "\r", "\r\n", "\n", "\t",
601 if ($search == NULL) {
602 $search = $_searchChars;
605 if ($replace == NULL) {
606 $replace = $_replaceChar;
609 return str_replace($search, $replace, $string);
614 * Use HTMLPurifier to clean up a text string and remove any potential
615 * xss attacks. This is primarily used in public facing pages which
616 * accept html as the input string
618 * @param string $string
621 * @return string the cleaned up string
624 public static function purifyHTML($string) {
625 static $_filter = null;
627 $config = HTMLPurifier_Config
::createDefault();
628 $config->set('Core.Encoding', 'UTF-8');
630 // Disable the cache entirely
631 $config->set('Cache.DefinitionImpl', null);
633 $_filter = new HTMLPurifier($config);
636 return $_filter->purify($string);
640 * Truncate $string; if $string exceeds $maxLen, place "..." at the end
642 * @param string $string
647 public static function ellipsify($string, $maxLen) {
648 $len = strlen($string);
649 if ($len <= $maxLen) {
653 return substr($string, 0, $maxLen-3) . '...';
658 * Generate a random string
664 public static function createRandom($len, $alphabet) {
665 $alphabetSize = strlen($alphabet);
667 for ($i = 0; $i < $len; $i++
) {
668 $result .= $alphabet{rand(1, $alphabetSize) - 1};
675 * "admin foo" => array(NULL,"admin foo")
676 * "cms:admin foo" => array("cms", "admin foo")
679 * @param string $string
680 * E.g. "view all contacts". Syntax: "[prefix:]name".
681 * @param null $defaultPrefix
683 * @return array (0 => string|NULL $prefix, 1 => string $value)
685 public static function parsePrefix($delim, $string, $defaultPrefix = NULL) {
686 $pos = strpos($string, $delim);
687 if ($pos === FALSE) {
688 return array($defaultPrefix, $string);
691 return array(substr($string, 0, $pos), substr($string, 1+
$pos));
696 * This function will mask part of the the user portion of an Email address (everything before the @)
698 * @param string $email
699 * The email address to be masked.
700 * @param string $maskChar
701 * The character used for masking.
702 * @param int $percent
703 * The percentage of the user portion to be masked.
705 * @return string returns the masked Email address
707 public static function maskEmail($email, $maskChar= '*', $percent=50) {
708 list($user, $domain) = preg_split("/@/", $email);
709 $len = strlen($user);
710 $maskCount = floor($len * $percent /100);
711 $offset = floor(($len - $maskCount) / 2);
713 $masked = substr($user, 0, $offset)
714 .str_repeat($maskChar, $maskCount)
715 .substr($user, $maskCount +
$offset);
717 return($masked.'@'.$domain);
721 * This function compares two strings
723 * @param string $strOne
725 * @param string $strTwo
728 * Boolean indicating whether you want the comparison to be case sensitive or not.
730 * @return boolean TRUE (string are identical); FALSE (strings are not identical)
732 public static function compareStr($strOne, $strTwo, $case) {
734 // Convert to lowercase and trim white spaces
735 if (strtolower(trim($strOne)) == strtolower(trim($strTwo))) {
736 // yes - they are identical
744 if ($case == FALSE) {
746 if (trim($strOne) == trim($strTwo)) {
747 // yes - they are identical
758 * Many parts of the codebase have a convention of internally passing around
759 * HTML-encoded URLs. This effectively means that "&" is replaced by "&"
760 * (because most other odd characters are %-escaped in URLs; and %-escaped
761 * strings don't need any extra escaping in HTML).
764 * URL with HTML entities.
765 * @return string URL without HTML entities
767 public static function unstupifyUrl($htmlUrl) {
768 return str_replace('&', '&', $htmlUrl);