1 <?php
  2 /**
  3  * Library of conversion functions, implemented as static methods.
  4  *
  5  * The methods are all of the form (format)2(format), where the format is one of
  6  * 
  7  *  raw: A UTF8 string
  8  *  attr: A UTF8 string suitable for inclusion in an HTML attribute
  9  *  js: A UTF8 string suitable for inclusion in a double-quoted javascript string.
 10  * 
 11  *  array: A PHP associative array
 12  *  json: JavaScript object notation
 13  *
 14  *  html: HTML source suitable for use in a page or email
 15  *  text: Plain-text content, suitable for display to a user as-is, or insertion in a plaintext email.
 16  * 
 17  * Objects of type {@link ViewableData} can have an "escaping type",
 18  * which determines if they are automatically escaped before output by {@link SSViewer}. 
 19  * 
 20  * @package sapphire
 21  * @subpackage misc
 22  */
 23 class Convert {
 24     
 25     /**
 26      * Convert a value to be suitable for an XML attribute.
 27      * 
 28      * @param array|string $val String to escape, or array of strings
 29      * @return array|string
 30      */
 31     static function raw2att($val) {
 32         if(is_array($val)) {
 33             foreach($val as $k => $v) $val[$k] = self::raw2att($v);
 34             return $val;
 35         } else {
 36             return str_replace(array('&','"',"'",'<','>'), array('&amp;','&quot;','&#39;','&lt;','&gt;'), $val);
 37         }
 38     }
 39     
 40     /**
 41      * Convert a value to be suitable for an HTML attribute.
 42      * 
 43      * This is useful for converting human readable values into
 44      * a value suitable for an ID or NAME attribute.
 45      * 
 46      * @see http://www.w3.org/TR/REC-html40/types.html#type-cdata
 47      * @uses Convert::raw2att()
 48      * @param array|string $val String to escape, or array of strings
 49      * @return array|string
 50      */
 51     static function raw2htmlatt($val) {
 52         if(is_array($val)) {
 53             foreach($val as $k => $v) $val[$k] = self::raw2htmlatt($v);
 54             return $val;
 55         } else {
 56             $val = self::rus2lat($val);
 57             $val = self::raw2att($val);
 58             $val = preg_replace('/[^a-zA-Z0-9\-_]*/', '', $val);
 59             return $val;
 60         }
 61     }
 62     
 63     /**
 64      * Ensure that text is properly escaped for XML.
 65      * 
 66      * @see http://www.w3.org/TR/REC-xml/#dt-escape
 67      * @param array|string $val String to escape, or array of strings
 68      * @return array|string
 69      */
 70     static function raw2xml($val) {
 71         if(is_array($val)) {
 72             foreach($val as $k => $v) $val[$k] = self::raw2xml($v);
 73             return $val;
 74         } else {
 75             return str_replace(array('&','<','>',"\n",'"',"'"), array('&amp;','&lt;','&gt;','<br />','&quot;','&#39;'), $val);
 76         }
 77     }
 78     
 79     /**
 80      * Ensure that text is properly escaped for Javascript.
 81      *
 82      * @param array|string $val String to escape, or array of strings
 83      * @return array|string
 84      */
 85     static function raw2js($val) {
 86         if(is_array($val)) {
 87             foreach($val as $k => $v) $val[$k] = self::raw2js($v);
 88             return $val;
 89         } else {
 90             return str_replace(array("\\", '"', "\n", "\r", "'"), array("\\\\", '\"', '\n', '\r', "\\'"), $val);
 91         }
 92     }
 93     
 94     /**
 95      * Uses the PHP 5.2 native json_encode function if available,
 96      * otherwise falls back to the Services_JSON class.
 97      * 
 98      * @see http://pear.php.net/pepr/pepr-proposal-show.php?id=198
 99      * @uses Director::baseFolder()
100      * @uses Services_JSON
101      *
102      * @param mixed $val
103      * @return string JSON safe string
104      */
105     static function raw2json($val) {
106         if(function_exists('json_encode')) {
107             return json_encode($val);   
108         } else {
109             require_once(Director::baseFolder() . '/sapphire/thirdparty/json/JSON.php');
110             $json = new Services_JSON();
111             return $json->encode($val);
112         }
113     }
114     
115     static function raw2sql($val) {
116         if(is_array($val)) {
117             foreach($val as $k => $v) $val[$k] = self::raw2sql($v);
118             return $val;
119         } else {
120             return DB::getConn()->addslashes($val);
121         }
122     }
123 
124     /**
125      * Convert XML to raw text.
126      * @uses html2raw()
127      * @todo Currently &#xxx; entries are stripped; they should be converted
128      */
129     static function xml2raw($val) {
130         if(is_array($val)) {
131             foreach($val as $k => $v) $val[$k] = self::xml2raw($v);
132             return $val;
133         } else {
134             // More complex text needs to use html2raw instead
135             if(strpos($val,'<') !== false) return self::html2raw($val);
136             
137             $converted = str_replace(array('&amp;','&lt;','&gt;','&quot;','&apos;', '&#39;'), array('&','<','>','"',"'", "'"), $val);
138             $converted = ereg_replace('&#[0-9]+;', '', $converted);
139             return $converted;
140         }
141     }
142     
143     /**
144      * Convert an array into a JSON encoded string.
145      * 
146      * @see http://pear.php.net/pepr/pepr-proposal-show.php?id=198
147      * @uses Director::baseFolder()
148      * @uses Services_JSON
149      * 
150      * @param array $val Array to convert
151      * @return string JSON encoded string
152      */
153     static function array2json($val) {
154         if(function_exists('json_encode')) {
155             return json_encode($val);
156         } else {
157             require_once(Director::baseFolder() . '/sapphire/thirdparty/json/JSON.php');
158             $json = new Services_JSON();
159             return $json->encode($val);
160         }
161     }
162     
163     /**
164      * Convert a JSON encoded string into an object.
165      * 
166      * @see http://pear.php.net/pepr/pepr-proposal-show.php?id=198
167      * @uses Director::baseFolder()
168      * @uses Services_JSON
169      *
170      * @param string $val
171      * @return mixed JSON safe string
172      */
173     static function json2obj($val) {
174         require_once(Director::baseFolder() . '/sapphire/thirdparty/json/JSON.php');
175         $json = new Services_JSON();
176         return $json->decode($val);
177     }
178 
179     /**
180      * Convert a JSON string into an array.
181      * 
182      * @uses json2obj
183      * @param string $val JSON string to convert
184      * @return array|boolean
185      */
186     static function json2array($val) {
187         $json = self::json2obj($val);
188         if(!$json) return false;
189         
190         $arr = array();
191         foreach($json as $k => $v) {
192             $arr[$k] = $v;
193         }
194         
195         return $arr;
196     }
197     
198     /**
199      * @uses recursiveXMLToArray()
200      */
201     static function xml2array($val) {
202         $xml = new SimpleXMLElement($val);
203         return self::recursiveXMLToArray($xml);
204     }
205 
206     /**
207      * Function recursively run from {@link Convert::xml2array()}
208      * @uses SimpleXMLElement
209      */
210     protected static function recursiveXMLToArray($xml) {
211         if(is_object($xml) && get_class($xml) == 'SimpleXMLElement') {
212             $attributes = $xml->attributes();
213             foreach($attributes as $k => $v) {
214                 if($v) $a[$k] = (string) $v;
215             }
216             $x = $xml;
217             $xml = get_object_vars($xml);
218         }
219         if(is_array($xml)) {
220             if(count($xml) == 0) return (string) $x; // for CDATA
221             foreach($xml as $key => $value) {
222                 $r[$key] = self::recursiveXMLToArray($value);
223             }
224             if(isset($a)) $r['@'] = $a; // Attributes
225             return $r;
226         }
227         return (string) $xml;
228     }
229     
230     /**
231      * Create a link if the string is a valid URL
232      * @param string The string to linkify
233      * @return A link to the URL if string is a URL
234      */
235     static function linkIfMatch($string) {
236         if( preg_match( '/^[a-z+]+\:\/\/[a-zA-Z0-9$-_.+?&=!*\'()%]+$/', $string ) )
237             return "<a style=\"white-space: nowrap\" href=\"$string\">$string</a>";
238         else
239             return $string;
240     }
241     
242     /**
243      * Simple conversion of HTML to plaintext.
244      * 
245      * @param $data string
246      * @param $preserveLinks boolean
247      * @param $wordwrap array 
248      */
249     static function html2raw($data, $preserveLinks = false, $wordWrap = 0, $config = null) {
250         $defaultConfig = array(
251             'PreserveLinks' => false,
252             'ReplaceBoldAsterisk' => true,
253             'CompressWhitespace' => true,
254             'ReplaceImagesWithAlt' => true,
255         );
256         if(isset($config)) {
257             $config = array_merge($defaultConfig,$config);
258         } else {
259             $config = $defaultConfig;
260         }
261 
262         // sTRIp style and script
263         /* $data = eregi_replace("<style(^A-Za-z0-9>][^>]*)?>.*</style[^>]*>","", $data);*/
264         /* $data = eregi_replace("<script(^A-Za-z0-9>][^>]*)?>.*</script[^>]*>","", $data);*/
265         
266         $data = preg_replace("/<style(^A-Za-z0-9>][^>]*)?>.*?<\/style[^>]*>/i","", $data);
267         $data = preg_replace("/<script(^A-Za-z0-9>][^>]*)?>.*?<\/script[^>]*>/i","", $data);
268 
269         if($config['ReplaceBoldAsterisk']) {
270             $data = preg_replace('%<(strong|b)( [^>]*)?>|</(strong|b)>%i','*',$data);
271         }
272         
273         // Expand hyperlinks
274         if(!$preserveLinks && !$config['PreserveLinks']) {
275             $data = preg_replace_callback('/<a[^>]*href\s*=\s*"([^"]*)">(.*?)<\/a>/i', function ($m) { return Convert::html2raw($m[2]) . ' [' . $m[1] . ']'; }, $data);
276             $data = preg_replace_callback('/<a[^>]*href\s*=\s*([^ ]*)>(.*?)<\/a>/i', function ($m) { return Convert::html2raw($m[2]) . ' [' . $m[1] . ']'; }, $data);
277 /*
278             $data = preg_replace('/<a[^>]*href\s*=\s*"([^"]*)">(.*?)<\/a>/ie', "Convert::html2raw('\\2').' [\\1]'", $data);
279             $data = preg_replace('/<a[^>]*href\s*=\s*([^ ]*)>(.*?)<\/a>/ie', "Convert::html2raw('\\2').' [\\1]'", $data);
280 */
281         }
282     
283         // Replace images with their alt tags
284         if($config['ReplaceImagesWithAlt']) {
285             $data = eregi_replace('<img[^>]*alt *= *"([^"]*)"[^>]*>', ' \\1 ', $data);
286             $data = eregi_replace('<img[^>]*alt *= *([^ ]*)[^>]*>', ' \\1 ', $data);
287         }
288     
289         // Compress whitespace
290         if($config['CompressWhitespace']) {
291             $data = ereg_replace("[\n\r\t ]+", " ", $data);
292         }
293         
294         // Parse newline tags
295         $data = ereg_replace("[ \n\r\t]*<[Hh][1-6]([^A-Za-z0-9>][^>]*)?> *", "\n\n", $data);
296         $data = ereg_replace("[ \n\r\t]*<[Pp]([^A-Za-z0-9>][^>]*)?> *", "\n\n", $data);
297         $data = ereg_replace("[ \n\r\t]*<[UuOo][Ll]([^A-Za-z0-9>][^>]*)?> *", "\n\n", $data);
298         $data = ereg_replace("[ \n\r\t]*<[Dd][Ii][Vv]([^A-Za-z0-9>][^>]*)?> *", "\n\n", $data);
299         $data = ereg_replace("\n\n\n+","\n\n", $data);
300         
301         $data = ereg_replace("<[Ll][Ii]([^A-Za-z0-9>][^>]*)?> *", "\n", $data); // "\n* "
302         $data = ereg_replace("<[Bb][Rr]([^A-Za-z0-9>][^>]*)?> *", "\n", $data);
303         $data = ereg_replace("<[Tt][Rr]([^A-Za-z0-9>][^>]*)?> *", "\n", $data);
304         $data = ereg_replace("</[Tt][Dd]([^A-Za-z0-9>][^>]*)?> *", "    ", $data);
305         $data = preg_replace('/<\/p>/i', "\n\n", $data );
306         
307         // Replace HTML entities
308         //$data = preg_replace("/&#([0-9]+);/e", 'chr(\1)', $data);
309         //$data = str_replace(array("&lt;","&gt;","&amp;","&nbsp;"), array("<", ">", "&", " "), $data);
310         $data = html_entity_decode($data, ENT_COMPAT , 'UTF-8');
311         // Remove all tags (but optionally keep links)
312         
313         // strip_tags seemed to be restricting the length of the output
314         // arbitrarily. This essentially does the same thing.
315         if(!$preserveLinks && !$config['PreserveLinks']) {
316             $data = preg_replace('/<\/?[^>]*>/','', $data);
317         } else {
318             $data = strip_tags($data, '<a>');
319         }
320         return ($wordWrap) ? trim(wordwrap(trim($data), $wordWrap)) : trim($data);
321     }
322     
323     /**
324      * There are no real specifications on correctly encoding mailto-links,
325      * but this seems to be compatible with most of the user-agents.
326      * Does nearly the same as rawurlencode().
327      * Please only encode the values, not the whole url, e.g.
328      * "mailto:test@test.com?subject=" . Convert::raw2mailto($subject)
329      * 
330      * @param $data string
331      * @return string
332      * @see http://www.ietf.org/rfc/rfc1738.txt
333      */
334     static function raw2mailto($data) {
335         return str_ireplace(
336             array("\n",'?','=',' ','(',')','&','@','"','\'',';'),
337             array('%0A','%3F','%3D','%20','%28','%29','%26','%40','%22','%27','%3B'),
338             $data
339         );
340     }
341 
342     static function rus2lat($string) {
343         $string = iconv('UTF-8', 'CP1251//TRANSLIT//IGNORE', $string);
344         $rus = array("/ŕ/", "/á/", "/â/", "/ă/", "/ä/", "/ĺ/", "/¸/", "/ć/", "/ç/", "/č/", "/é/", "/ę/", "/ë/", "/ě/", "/í/", "/î/", "/ď/", "/đ/", "/ń/", "/ň/", "/ó/", "/ô/", "/ő/", "/ö/", "/÷/", "/ř/", "/ů/", "/ű/", "/ý/", "/ţ/", "/˙/", "/ü/", "/ú/", "/ż/", "/ş/", 
345             "/Ŕ/", "/Á/", "/Â/", "/Ă/", "/Ä/", "/Ĺ/", "/¨/", "/Ć/", "/Ç/", "/Č/", "/É/", "/Ę/", "/Ë/", "/Ě/", "/Í/", "/Î/", "/Ď/", "/Đ/", "/Ń/", "/Ň/", "/Ó/", "/Ô/", "/Ő/", "/Ö/", "/×/", "/Ř/", "/Ů/", "/Ű/", "/Ý/", "/Ţ/", "/ß/", "/Ü/", "/Ú/", "/Ż/", "/Ş/"
346         );
347         
348         $lat = array("a", "b", "v", "g", "d", "e", "e", "zh", "z", "i", "j", "k", "l", "m", "n", "o", "p", "r", "s", "t", "u", "f", "h", "c", "ch", "sh", "sh'", "y", "e", "yu", "ya", "'", "'", "i", "e",
349             "A", "B", "V", "G", "D", "E", "E", "ZH", "Z", "I", "J", "K", "L", "M", "N", "O", "P", "R", "S", "T", "U", "F", "H", "C", "CH", "SH", "SH'", "Y", "E", "YU", "YA", "'", "'", "I", "E"
350         );
351         return preg_replace($rus, $lat, $string);
352     }
353 
354     /**
355      * Generate correct name of object for integer number of object
356      *
357      * @param $num      int     number of object
358      * @param $name1    string  name for *[1] object
359      * @param $name2_4  string  name for *[2-4] objects
360      * @param $nameOther    string  name for 11-19, *[0,5-9] objects
361      * @return string
362      */
363     static function number2name($num, $name1, $name2_4, $nameOther) {
364         if ($num >= 5 && $num <= 20) return $nameOther;
365         $tail = $num % 100;
366         $num = $num % 10;
367         if ($num == 1) return $name1;
368         if ($num >= 2 && $num <= 4 && ($tail < 11 || $tail > 19)) return $name2_4;
369         return $nameOther;
370     }
371 }
372 
373