CRM/Utils/String.php

   1 <?php
   2 /*
   3  +--------------------------------------------------------------------+
   4  | CiviCRM version 4.5                                                |
   5  +--------------------------------------------------------------------+
   6  | Copyright CiviCRM LLC (c) 2004-2014                                |
   7  +--------------------------------------------------------------------+
   8  | This file is a part of CiviCRM.                                    |
   9  |                                                                    |
  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.   |
  13  |                                                                    |
  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.        |
  18  |                                                                    |
  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  +--------------------------------------------------------------------+
  26 */
  27
  28 /**
  29  *
  30  * @package CRM
  31  * @copyright CiviCRM LLC (c) 2004-2014
  32  * $Id$
  33  *
  34  */
  35
  36 require_once 'HTML/QuickForm/Rule/Email.php';
  37
  38 /**
  39  * This class contains string functions
  40  *
  41  */
  42 class CRM_Utils_String {
  43   CONST COMMA = ",", SEMICOLON = ";", SPACE = " ", TAB = "\t", LINEFEED = "\n", CARRIAGELINE = "\r\n", LINECARRIAGE = "\n\r", CARRIAGERETURN = "\r";
  44
  45   /**
  46    * List of all letters and numbers
  47    */
  48   const ALPHANUMERIC = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890';
  49
  50   /**
  51    * Convert a display name into a potential variable
  52    * name that we could use in forms/code
  53    *
  54    * @param  name    Name of the string
  55    *
  56    * @param int $maxLength
  57    *
  58    * @return string  An equivalent variable name
  59    *
  60    * @access public
  61    *
  62    * @return string (or null)
  63    * @static
  64    */
  65   static function titleToVar($title, $maxLength = 31) {
  66     $variable = self::munge($title, '_', $maxLength);
  67
  68     if (CRM_Utils_Rule::title($variable, $maxLength)) {
  69       return $variable;
  70     }
  71
  72     // if longer than the maxLength lets just return a substr of the
  73     // md5 to prevent errors downstream
  74     return substr(md5($title), 0, $maxLength);
  75   }
  76
  77   /**
  78    * given a string, replace all non alpha numeric characters and
  79    * spaces with the replacement character
  80    *
  81    * @param string $name the name to be worked on
  82    * @param string $char the character to use for non-valid chars
  83    * @param int    $len  length of valid variables
  84    *
  85    * @access public
  86    *
  87    * @return string returns the manipulated string
  88    * @static
  89    */
  90   static function munge($name, $char = '_', $len = 63) {
  91     // replace all white space and non-alpha numeric with $char
  92     // we only use the ascii character set since mysql does not create table names / field names otherwise
  93     // CRM-11744
  94     $name = preg_replace('/[^a-zA-Z0-9]+/', $char, trim($name));
  95
  96     if ($len) {
  97       // lets keep variable names short
  98       return substr($name, 0, $len);
  99     }
 100     else {
 101       return $name;
 102     }
 103   }
 104
 105   /**
 106    *
 107    * Takes a variable name and munges it randomly into another variable name
 108    *
 109    * @param  string $name    Initial Variable Name
 110    * @param int     $len  length of valid variables
 111    *
 112    * @return string  Randomized Variable Name
 113    * @access public
 114    * @static
 115    */
 116   static function rename($name, $len = 4) {
 117     $rand = substr(uniqid(), 0, $len);
 118     return substr_replace($name, $rand, -$len, $len);
 119   }
 120
 121   /**
 122    * takes a string and returns the last tuple of the string.
 123    * useful while converting file names to class names etc
 124    *
 125    * @param string $string the input string
 126    * @param \char|string $char $char   the character used to demarcate the componets
 127    *
 128    * @access public
 129    *
 130    * @return string the last component
 131    * @static
 132    */
 133   static function getClassName($string, $char = '_') {
 134     $names = array();
 135     if (!is_array($string)) {
 136       $names = explode($char, $string);
 137     }
 138     if (!empty($names)) {
 139       return array_pop($names);
 140     }
 141   }
 142
 143   /**
 144    * appends a name to a string and seperated by delimiter.
 145    * does the right thing for an empty string
 146    *
 147    * @param string $str   the string to be appended to
 148    * @param string $delim the delimiter to use
 149    * @param mixed  $name  the string (or array of strings) to append
 150    *
 151    * @return void
 152    * @access public
 153    * @static
 154    */
 155   static function append(&$str, $delim, $name) {
 156     if (empty($name)) {
 157       return;
 158     }
 159
 160     if (is_array($name)) {
 161       foreach ($name as $n) {
 162         if (empty($n)) {
 163           continue;
 164         }
 165         if (empty($str)) {
 166           $str = $n;
 167         }
 168         else {
 169           $str .= $delim . $n;
 170         }
 171       }
 172     }
 173     else {
 174       if (empty($str)) {
 175         $str = $name;
 176       }
 177       else {
 178         $str .= $delim . $name;
 179       }
 180     }
 181   }
 182
 183   /**
 184    * determine if the string is composed only of ascii characters
 185    *
 186    * @param string  $str input string
 187    * @param boolean $utf8 attempt utf8 match on failure (default yes)
 188    *
 189    * @return boolean    true if string is ascii
 190    * @access public
 191    * @static
 192    */
 193   static function isAscii($str, $utf8 = TRUE) {
 194     if (!function_exists('mb_detect_encoding')) {
 195       // eliminate all white space from the string
 196       $str = preg_replace('/\s+/', '', $str);
 197       // FIXME:  This is a pretty brutal hack to make utf8 and 8859-1 work.
 198
 199       /* match low- or high-ascii characters */
 200       if (preg_match('/[\x00-\x20]|[\x7F-\xFF]/', $str)) {
 201         // || // low ascii characters
 202         // high ascii characters
 203         //  preg_match( '/[\x7F-\xFF]/', $str ) ) {
 204         if ($utf8) {
 205           /* if we did match, try for utf-8, or iso8859-1 */
 206
 207           return self::isUtf8($str);
 208         }
 209         else {
 210           return FALSE;
 211         }
 212       }
 213       return TRUE;
 214     }
 215     else {
 216       $order = array('ASCII');
 217       if ($utf8) {
 218         $order[] = 'UTF-8';
 219       }
 220       $enc = mb_detect_encoding($str, $order, TRUE);
 221       return ($enc == 'ASCII' || $enc == 'UTF-8');
 222     }
 223   }
 224
 225   /**
 226    * determine the string replacements for redaction
 227    * on the basis of the regular expressions
 228    *
 229    * @param string $str        input string
 230    * @param array  $regexRules regular expression to be matched w/ replacements
 231    *
 232    * @return array $match      array of strings w/ corresponding redacted outputs
 233    * @access public
 234    * @static
 235    */
 236   static function regex($str, $regexRules) {
 237     //redact the regular expressions
 238     if (!empty($regexRules) && isset($str)) {
 239       static $matches, $totalMatches, $match = array();
 240       foreach ($regexRules as $pattern => $replacement) {
 241         preg_match_all($pattern, $str, $matches);
 242         if (!empty($matches[0])) {
 243           if (empty($totalMatches)) {
 244             $totalMatches = $matches[0];
 245           }
 246           else {
 247             $totalMatches = array_merge($totalMatches, $matches[0]);
 248           }
 249           $match = array_flip($totalMatches);
 250         }
 251       }
 252     }
 253
 254     if (!empty($match)) {
 255       foreach ($match as $matchKey => & $dontCare) {
 256         foreach ($regexRules as $pattern => $replacement) {
 257           if (preg_match($pattern, $matchKey)) {
 258             $dontCare = $replacement . substr(md5($matchKey), 0, 5);
 259             break;
 260           }
 261         }
 262       }
 263       return $match;
 264     }
 265     return CRM_Core_DAO::$_nullArray;
 266   }
 267
 268   static function redaction($str, $stringRules) {
 269     //redact the strings
 270     if (!empty($stringRules)) {
 271       foreach ($stringRules as $match => $replace) {
 272         $str = str_ireplace($match, $replace, $str);
 273       }
 274     }
 275
 276     //return the redacted output
 277     return $str;
 278   }
 279
 280   /**
 281    * Determine if a string is composed only of utf8 characters
 282    *
 283    * @param string $str  input string
 284    * @access public
 285    * @static
 286    *
 287    * @return boolean
 288    */
 289   static function isUtf8($str) {
 290     if (!function_exists(mb_detect_encoding)) {
 291       // eliminate all white space from the string
 292       $str = preg_replace('/\s+/', '', $str);
 293
 294       /* pattern stolen from the php.net function documentation for
 295              * utf8decode();
 296              * comment by JF Sebastian, 30-Mar-2005
 297              */
 298
 299       return preg_match('/^([\x00-\x7f]|[\xc2-\xdf][\x80-\xbf]|\xe0[\xa0-\xbf][\x80-\xbf]|[\xe1-\xec][\x80-\xbf]{2}|\xed[\x80-\x9f][\x80-\xbf]|[\xee-\xef][\x80-\xbf]{2}|f0[\x90-\xbf][\x80-\xbf]{2}|[\xf1-\xf3][\x80-\xbf]{3}|\xf4[\x80-\x8f][\x80-\xbf]{2})*$/', $str);
 300       // ||
 301       // iconv('ISO-8859-1', 'UTF-8', $str);
 302     }
 303     else {
 304       $enc = mb_detect_encoding($str, array('UTF-8'), TRUE);
 305       return ($enc !== FALSE);
 306     }
 307   }
 308
 309   /**
 310    * determine if two href's are equivalent (fuzzy match)
 311    *
 312    * @param string $url1 the first url to be matched
 313    * @param string $url2 the second url to be matched against
 314    *
 315    * @return boolean true if the urls match, else false
 316    * @access public
 317    * @static
 318    */
 319   static function match($url1, $url2) {
 320     $url1 = strtolower($url1);
 321     $url2 = strtolower($url2);
 322
 323     $url1Str = parse_url($url1);
 324     $url2Str = parse_url($url2);
 325
 326     if ($url1Str['path'] == $url2Str['path'] &&
 327       self::extractURLVarValue(CRM_Utils_Array::value('query', $url1Str)) == self::extractURLVarValue(CRM_Utils_Array::value('query', $url2Str))
 328     ) {
 329       return TRUE;
 330     }
 331     return FALSE;
 332   }
 333
 334   /**
 335    * Function to extract variable values
 336    *
 337    * @param  mix $query this is basically url
 338    *
 339    * @return mix $v  returns civicrm url (eg: civicrm/contact/search/...)
 340    * @access public
 341    * @static
 342    */
 343   static function extractURLVarValue($query) {
 344     $config = CRM_Core_Config::singleton();
 345     $urlVar = $config->userFrameworkURLVar;
 346
 347     $params = explode('&', $query);
 348     foreach ($params as $p) {
 349       if (strpos($p, '=')) {
 350         list($k, $v) = explode('=', $p);
 351         if ($k == $urlVar) {
 352           return $v;
 353         }
 354       }
 355     }
 356     return NULL;
 357   }
 358
 359   /**
 360    * translate a true/false/yes/no string to a 0 or 1 value
 361    *
 362    * @param string $str  the string to be translated
 363    *
 364    * @return boolean
 365    * @access public
 366    * @static
 367    */
 368   static function strtobool($str) {
 369     if (!is_scalar($str)) {
 370       return FALSE;
 371     }
 372
 373     if (preg_match('/^(y(es)?|t(rue)?|1)$/i', $str)) {
 374       return TRUE;
 375     }
 376     return FALSE;
 377   }
 378
 379   /**
 380    * returns string '1' for a true/yes/1 string, and '0' for no/false/0 else returns false
 381    *
 382    * @param string $str  the string to be translated
 383    *
 384    * @return boolean
 385    * @access public
 386    * @static
 387    */
 388   static function strtoboolstr($str) {
 389     if (!is_scalar($str)) {
 390       return FALSE;
 391     }
 392
 393     if (preg_match('/^(y(es)?|t(rue)?|1)$/i', $str)) {
 394       return '1';
 395     }
 396     elseif (preg_match('/^(n(o)?|f(alse)?|0)$/i', $str)) {
 397       return '0';
 398     }
 399     else {
 400       return FALSE;
 401     }
 402   }
 403
 404   /**
 405    * Convert a HTML string into a text one using html2text
 406    *
 407    * @param string $html  the string to be converted
 408    *
 409    * @return string       the converted string
 410    * @access public
 411    * @static
 412    */
 413   static function htmlToText($html) {
 414     require_once 'packages/html2text/rcube_html2text.php';
 415     $token_html = preg_replace('!\{([a-z_.]+)\}!i', 'token:{$1}', $html);
 416     $converter = new rcube_html2text($token_html);
 417     $token_text = $converter->get_text();
 418     $text = preg_replace('!token\:\{([a-z_.]+)\}!i', '{$1}', $token_text);
 419     return $text;
 420   }
 421
 422   static function extractName($string, &$params) {
 423     $name = trim($string);
 424     if (empty($name)) {
 425       return;
 426     }
 427
 428     // strip out quotes
 429     $name = str_replace('"', '', $name);
 430     $name = str_replace('\'', '', $name);
 431
 432     // check for comma in name
 433     if (strpos($name, ',') !== FALSE) {
 434
 435       // name has a comma - assume lname, fname [mname]
 436       $names = explode(',', $name);
 437       if (count($names) > 1) {
 438         $params['last_name'] = trim($names[0]);
 439
 440         // check for space delim
 441         $fnames = explode(' ', trim($names[1]));
 442         if (count($fnames) > 1) {
 443           $params['first_name'] = trim($fnames[0]);
 444           $params['middle_name'] = trim($fnames[1]);
 445         }
 446         else {
 447           $params['first_name'] = trim($fnames[0]);
 448         }
 449       }
 450       else {
 451         $params['first_name'] = trim($names[0]);
 452       }
 453     }
 454     else {
 455       // name has no comma - assume fname [mname] fname
 456       $names = explode(' ', $name);
 457       if (count($names) == 1) {
 458         $params['first_name'] = $names[0];
 459       }
 460       elseif (count($names) == 2) {
 461         $params['first_name'] = $names[0];
 462         $params['last_name'] = $names[1];
 463       }
 464       else {
 465         $params['first_name'] = $names[0];
 466         $params['middle_name'] = $names[1];
 467         $params['last_name'] = $names[2];
 468       }
 469     }
 470   }
 471
 472   static function &makeArray($string) {
 473     $string = trim($string);
 474
 475     $values = explode("\n", $string);
 476     $result = array();
 477     foreach ($values as $value) {
 478       list($n, $v) = CRM_Utils_System::explode('=', $value, 2);
 479       if (!empty($v)) {
 480         $result[trim($n)] = trim($v);
 481       }
 482     }
 483     return $result;
 484   }
 485
 486   /**
 487    * Given an ezComponents-parsed representation of
 488    * a text with alternatives return only the first one
 489    *
 490    * @param string $full  all alternatives as a long string (or some other text)
 491    *
 492    * @return string       only the first alternative found (or the text without alternatives)
 493    */
 494   static function stripAlternatives($full) {
 495     $matches = array();
 496     preg_match('/-ALTERNATIVE ITEM 0-(.*?)-ALTERNATIVE ITEM 1-.*-ALTERNATIVE END-/s', $full, $matches);
 497
 498     if (isset($matches[1]) &&
 499       trim(strip_tags($matches[1])) != ''
 500     ) {
 501       return $matches[1];
 502     }
 503     else {
 504       return $full;
 505     }
 506   }
 507
 508   /**
 509    * strip leading, trailing, double spaces from string
 510    * used for postal/greeting/addressee
 511    *
 512    * @param string  $string input string to be cleaned
 513    *
 514    * @return string the cleaned string
 515    * @access public
 516    * @static
 517    */
 518   static function stripSpaces($string) {
 519     return (empty($string)) ? $string : preg_replace("/\s{2,}/", " ", trim($string));
 520   }
 521
 522   /**
 523    * This function is used to clean the URL 'path' variable that we use
 524    * to construct CiviCRM urls by removing characters from the path variable
 525    *
 526    * @param string $string  the input string to be sanitized
 527    * @param array  $search  the characters to be sanitized
 528    * @param string $replace the character to replace it with
 529    *
 530    * @return string the sanitized string
 531    * @access public
 532    * @static
 533    */
 534   static function stripPathChars($string,
 535     $search = NULL,
 536     $replace = NULL
 537   ) {
 538     static $_searchChars = NULL;
 539     static $_replaceChar = NULL;
 540
 541     if (empty($string)) {
 542       return $string;
 543     }
 544
 545     if ($_searchChars == NULL) {
 546       $_searchChars = array(
 547         '&', ';', ',', '=', '$',
 548         '"', "'", '\\',
 549         '<', '>', '(', ')',
 550         ' ', "\r", "\r\n", "\n", "\t",
 551       );
 552       $_replaceChar = '_';
 553     }
 554
 555
 556     if ($search == NULL) {
 557       $search = $_searchChars;
 558     }
 559
 560     if ($replace == NULL) {
 561       $replace = $_replaceChar;
 562     }
 563
 564     return str_replace($search, $replace, $string);
 565   }
 566
 567
 568   /**
 569    * Use HTMLPurifier to clean up a text string and remove any potential
 570    * xss attacks. This is primarily used in public facing pages which
 571    * accept html as the input string
 572    *
 573    * @param string $string the input string
 574    *
 575    * @return string the cleaned up string
 576    * @public
 577    * @static
 578    */
 579   static function purifyHTML($string) {
 580     static $_filter = null;
 581     if (!$_filter) {
 582       $config = HTMLPurifier_Config::createDefault();
 583       $config->set('Core.Encoding', 'UTF-8');
 584
 585       // Disable the cache entirely
 586       $config->set('Cache.DefinitionImpl', null);
 587
 588       $_filter = new HTMLPurifier($config);
 589     }
 590
 591     return $_filter->purify($string);
 592   }
 593
 594   /**
 595    * Truncate $string; if $string exceeds $maxLen, place "..." at the end
 596    *
 597    * @param string $string
 598    * @param int $maxLen
 599    *
 600    * @return string
 601    */
 602   static function ellipsify($string, $maxLen) {
 603     $len = strlen($string);
 604     if ($len <= $maxLen) {
 605       return $string;
 606     }
 607     else {
 608       return substr($string, 0, $maxLen-3) . '...';
 609     }
 610   }
 611
 612   /**
 613    * Generate a random string
 614    *
 615    * @param $len
 616    * @param $alphabet
 617    * @return string
 618    */
 619   public static function createRandom($len, $alphabet) {
 620     $alphabetSize = strlen($alphabet);
 621     $result = '';
 622     for ($i = 0; $i < $len; $i++) {
 623       $result .= $alphabet{rand(1, $alphabetSize) - 1};
 624     }
 625     return $result;
 626   }
 627
 628   /**
 629    * Examples:
 630    * "admin foo" => array(NULL,"admin foo")
 631    * "cms:admin foo" => array("cms", "admin foo")
 632    *
 633    * @param $delim
 634    * @param string $string e.g. "view all contacts". Syntax: "[prefix:]name"
 635    * @param null $defaultPrefix
 636    *
 637    * @return array (0 => string|NULL $prefix, 1 => string $value)
 638    */
 639   public static function parsePrefix($delim, $string, $defaultPrefix = NULL) {
 640     $pos = strpos($string, $delim);
 641     if ($pos === FALSE) {
 642       return array($defaultPrefix, $string);
 643     }
 644     else {
 645       return array(substr($string, 0, $pos), substr($string, 1+$pos));
 646     }
 647   }
 648
 649   /**
 650    * this function will mask part of the the user portion of an Email address (everything before the @)
 651    *
 652    * @param string $email the email address to be masked
 653    * @param string $maskChar the character used for masking
 654    * @param integer $percent the percentage of the user portion to be masked
 655    *
 656    * @return string returns the masked Email address
 657    */
 658   public static function maskEmail($email, $maskChar= '*', $percent=50) {
 659     list($user, $domain) = preg_split("/@/", $email);
 660     $len = strlen($user);
 661     $maskCount = floor($len * $percent /100);
 662     $offset = floor(($len - $maskCount) / 2);
 663
 664     $masked = substr($user, 0, $offset)
 665       .str_repeat($maskChar, $maskCount)
 666       .substr($user, $maskCount + $offset);
 667
 668     return($masked.'@'.$domain);
 669   }
 670
 671   /**
 672    * this function compares two strings
 673    *
 674    * @param string $strOne string one
 675    * @param string $strTwo string two
 676    * @param boolean $case boolean indicating whether you want the comparison to be case sensitive or not
 677    *
 678    * @return boolean TRUE (string are identical); FALSE (strings are not identical)
 679    */
 680   public static function compareStr($strOne, $strTwo, $case) {
 681     if ($case == TRUE) {
 682       // Convert to lowercase and trim white spaces
 683       if (strtolower(trim($strOne)) == strtolower(trim($strTwo))) {
 684         // yes - they are identical
 685         return TRUE;
 686       }
 687       else {
 688         // not identical
 689         return FALSE;
 690       }
 691     }
 692     if ($case == FALSE) {
 693       // Trim white spaces
 694       if (trim($strOne) == trim($strTwo)) {
 695         // yes - they are identical
 696         return TRUE;
 697       }
 698       else {
 699         // not identical
 700         return FALSE;
 701       }
 702     }
 703   }
 704
 705 }
 706