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());
12 /// @todo add to iso table the characters from cp_1252 range, i.e. 128 to 159?
13 /// These will NOT be present in true ISO-8859-1, but will save the unwary windows user from sending junk
14 /// (though no luck when receiving them...)
16 protected $xml_cp1252_Entities = array('in' => array(), out' => array(
17 '€', '?', '‚', 'ƒ',
18 '„', '…', '†', '‡',
19 'ˆ', '‰', 'Š', '‹',
20 'Œ', '?', 'Ž', '?',
21 '?', '‘', '’', '“',
22 '”', '•', '–', '—',
23 '˜', '™', 'š', '›',
24 'œ', '?', 'ž', 'Ÿ'
28 protected $charset_supersets = array(
29 'US-ASCII' => array('ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4',
30 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8',
31 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-12',
32 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'UTF-8',
33 'EUC-JP', 'EUC-', 'EUC-KR', 'EUC-CN',),
36 protected static $instance = null;
39 * This class is singleton for performance reasons.
43 public static function instance()
45 if (self::$instance === null) {
46 self::$instance = new self();
49 return self::$instance;
52 private function __construct()
54 for ($i = 0; $i < 32; $i++) {
55 $this->xml_iso88591_Entities["in"][] = chr($i);
56 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
59 for ($i = 160; $i < 256; $i++) {
60 $this->xml_iso88591_Entities["in"][] = chr($i);
61 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
64 /*for ($i = 128; $i < 160; $i++)
66 $this->xml_cp1252_Entities['in'][] = chr($i);
71 * Convert a string to the correct XML representation in a target charset.
73 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
74 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
75 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
76 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
77 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
78 * bound by RFC 3023 to emit strict us-ascii.
80 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
81 * @todo make usage of iconv() or recode_string() or mb_string() where available
84 * @param string $srcEncoding
85 * @param string $destEncoding
89 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
91 if ($srcEncoding == '') {
92 // lame, but we know no better...
93 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
96 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
97 switch ($conversion) {
99 case 'ISO-8859-1_US-ASCII':
100 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
101 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
104 case 'ISO-8859-1_UTF-8':
105 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
106 $escapedData = utf8_encode($escapedData);
109 case 'ISO-8859-1_ISO-8859-1':
110 case 'US-ASCII_US-ASCII':
111 case 'US-ASCII_UTF-8':
113 case 'US-ASCII_ISO-8859-1':
115 //case 'CP1252_CP1252':
116 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
120 case 'UTF-8_US-ASCII':
121 case 'UTF-8_ISO-8859-1':
122 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
124 // be kind to users creating string xmlrpc values out of different php types
125 $data = (string)$data;
127 for ($nn = 0; $nn < $ns; $nn++) {
130 // 7 bits: 0bbbbbbb (127)
132 /// @todo shall we replace this with a (supposedly) faster str_replace?
135 $escapedData .= '"';
138 $escapedData .= '&';
141 $escapedData .= ''';
144 $escapedData .= '<';
147 $escapedData .= '>';
152 } // 11 bits: 110bbbbb 10bbbbbb (2047)
153 elseif ($ii >> 5 == 6) {
155 $ii = ord($data[$nn + 1]);
157 $ii = ($b1 * 64) + $b2;
158 $ent = sprintf('&#%d;', $ii);
159 $escapedData .= $ent;
161 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
162 elseif ($ii >> 4 == 14) {
164 $ii = ord($data[$nn + 1]);
166 $ii = ord($data[$nn + 2]);
168 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
169 $ent = sprintf('&#%d;', $ii);
170 $escapedData .= $ent;
172 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
173 elseif ($ii >> 3 == 30) {
175 $ii = ord($data[$nn + 1]);
177 $ii = ord($data[$nn + 2]);
179 $ii = ord($data[$nn + 3]);
181 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
182 $ent = sprintf('&#%d;', $ii);
183 $escapedData .= $ent;
188 // when converting to latin-1, do not be so eager with using entities for characters 160-255
189 if ($conversion == 'UTF-8_ISO-8859-1') {
190 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
196 case 'CP1252_US-ASCII':
197 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
198 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
199 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
202 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
203 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all allone will NOT convert them)
204 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
205 $escapedData = utf8_encode($escapedData);
207 case 'CP1252_ISO-8859-1':
208 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
209 // 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...
210 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
216 Logger::instance()->errorLog('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
223 * Checks if a given charset encoding is present in a list of encodings or if it is a valid subset of any encoding
226 * @param string $encoding charset to be tested
227 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
231 public function isValidCharset($encoding, $validList)
233 if (is_string($validList)) {
234 $validList = explode(',', $validList);
236 if (@in_array(strtoupper($encoding), $validList)) {
239 if (array_key_exists($encoding, $this->charset_supersets)) {
240 foreach ($validList as $allowed) {
241 if (in_array($allowed, $this->charset_supersets[$encoding])) {
252 * Used only for backwards compatibility
255 * @param string $charset
259 * @throws \Exception for unknown/unsupported charsets
261 public function getEntities($charset)
263 //trigger_error('Method ' . __METHOD__ . ' is deprecated', E_USER_DEPRECATED);
268 return $this->xml_iso88591_Entities;
270 throw new \Exception('Unsupported charset: ' . $charset);