3 namespace PhpXmlRpc\Helper;
5 use PhpXmlRpc\PhpXmlRpc;
9 // tables used for transcoding different charsets into us-ascii xml
10 protected $xml_iso88591_Entities = array("in" => array(), "out" => array());
11 protected $xml_iso88591_utf8 = array("in" => array(), "out" => array());
13 /// @todo add to iso table the characters from cp_1252 range, i.e. 128 to 159?
14 /// These will NOT be present in true ISO-8859-1, but will save the unwary
15 /// windows user from sending junk (though no luck when receiving them...)
17 protected $xml_cp1252_Entities = array('in' => array(), out' => array(
18 '€', '?', '‚', 'ƒ',
19 '„', '…', '†', '‡',
20 'ˆ', '‰', 'Š', '‹',
21 'Œ', '?', 'Ž', '?',
22 '?', '‘', '’', '“',
23 '”', '•', '–', '—',
24 '˜', '™', 'š', '›',
25 'œ', '?', 'ž', 'Ÿ'
29 protected $charset_supersets = array(
30 'US-ASCII' => array('ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4',
31 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8',
32 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-12',
33 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'UTF-8',
34 'EUC-JP', 'EUC-', 'EUC-KR', 'EUC-CN',),
37 protected static $instance = null;
40 * This class is singleton for performance reasons.
44 public static function instance()
46 if (self::$instance === null) {
47 self::$instance = new self();
50 return self::$instance;
53 private function __construct()
55 for ($i = 0; $i < 32; $i++) {
56 $this->xml_iso88591_Entities["in"][] = chr($i);
57 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
60 for ($i = 160; $i < 256; $i++) {
61 $this->xml_iso88591_Entities["in"][] = chr($i);
62 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
65 /*for ($i = 128; $i < 160; $i++)
67 $this->xml_cp1252_Entities['in'][] = chr($i);
72 * Convert a string to the correct XML representation in a target charset.
74 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
75 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
76 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
77 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
78 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
79 * bound by RFC 3023 to emit strict us-ascii.
81 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
82 * @todo make usage of iconv() or recode_string() or mb_string() where available
85 * @param string $srcEncoding
86 * @param string $destEncoding
90 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
92 if ($srcEncoding == '') {
93 // lame, but we know no better...
94 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
97 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
98 switch ($conversion) {
100 case 'ISO-8859-1_US-ASCII':
101 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
102 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
105 case 'ISO-8859-1_UTF-8':
106 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
107 $escapedData = utf8_encode($escapedData);
110 case 'ISO-8859-1_ISO-8859-1':
111 case 'US-ASCII_US-ASCII':
112 case 'US-ASCII_UTF-8':
114 case 'US-ASCII_ISO-8859-1':
116 //case 'CP1252_CP1252':
117 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
121 case 'UTF-8_US-ASCII':
122 case 'UTF-8_ISO-8859-1':
123 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
125 // be kind to users creating string xmlrpc values out of different php types
126 $data = (string)$data;
128 for ($nn = 0; $nn < $ns; $nn++) {
131 // 7 bits: 0bbbbbbb (127)
133 /// @todo shall we replace this with a (supposedly) faster str_replace?
136 $escapedData .= '"';
139 $escapedData .= '&';
142 $escapedData .= ''';
145 $escapedData .= '<';
148 $escapedData .= '>';
153 } // 11 bits: 110bbbbb 10bbbbbb (2047)
154 elseif ($ii >> 5 == 6) {
156 $ii = ord($data[$nn + 1]);
158 $ii = ($b1 * 64) + $b2;
159 $ent = sprintf('&#%d;', $ii);
160 $escapedData .= $ent;
162 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
163 elseif ($ii >> 4 == 14) {
165 $ii = ord($data[$nn + 1]);
167 $ii = ord($data[$nn + 2]);
169 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
170 $ent = sprintf('&#%d;', $ii);
171 $escapedData .= $ent;
173 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
174 elseif ($ii >> 3 == 30) {
176 $ii = ord($data[$nn + 1]);
178 $ii = ord($data[$nn + 2]);
180 $ii = ord($data[$nn + 3]);
182 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
183 $ent = sprintf('&#%d;', $ii);
184 $escapedData .= $ent;
189 // when converting to latin-1, do not be so eager with using entities for characters 160-255
190 if ($conversion == 'UTF-8_ISO-8859-1') {
191 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
197 case 'CP1252_US-ASCII':
198 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
199 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
200 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
203 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
204 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all allone will NOT convert them)
205 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
206 $escapedData = utf8_encode($escapedData);
208 case 'CP1252_ISO-8859-1':
209 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
210 // we might as well replace all funky chars with a '?' here, but we are kind and leave it to the receiving application layer to decide what to do with these weird entities...
211 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
217 error_log('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
224 * Checks if a given charset encoding is present in a list of encodings or
225 * if it is a valid subset of any encoding in the list.
227 * @param string $encoding charset to be tested
228 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
232 public function isValidCharset($encoding, $validList)
234 if (is_string($validList)) {
235 $validList = explode(',', $validList);
237 if (@in_array(strtoupper($encoding), $validList)) {
240 if (array_key_exists($encoding, $this->charset_supersets)) {
241 foreach ($validList as $allowed) {
242 if (in_array($allowed, $this->charset_supersets[$encoding])) {
253 * Used only for backwards compatibility
256 * @param string $charset
260 * @throws \Exception for unknown/unsupported charsets
262 public function getEntities($charset)
267 return $this->xml_iso88591_Entities;
269 throw new \Exception('Unsupported charset: ' . $charset);