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 should we add to the latin-1 table the characters from cp_1252 range, i.e. 128 to 159 ?
13 /// Those 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...)
15 /// Note also that, apparently, while 'ISO/IEC 8859-1' has no characters defined for bytes 128 to 159,
16 /// IANA ISO-8859-1 does have well-defined 'C1' control codes for those - wikipedia's page on latin-1 says:
17 /// "ISO-8859-1 is the IANA preferred name for this standard when supplemented with the C0 and C1 control codes from ISO/IEC 6429."
18 /// Check what mbstring/iconv do by default with those?
20 //protected $xml_cp1252_Entities = array('in' => array(), out' => array());
22 protected $charset_supersets = array(
23 'US-ASCII' => array('ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4',
24 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8',
25 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-12',
26 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'UTF-8',
27 'EUC-JP', 'EUC-', 'EUC-KR', 'EUC-CN',),
30 protected static $instance = null;
33 * This class is singleton for performance reasons.
34 * @todo can't we just make $xml_iso88591_Entities a static variable instead ?
38 public static function instance()
40 if (self::$instance === null) {
41 self::$instance = new self();
44 return self::$instance;
48 * Force usage as singleton
50 protected function __construct()
55 * @param string $tableName
56 * @throws \Exception for unsupported $tableName
58 protected function buildConversionTable($tableName)
61 case 'xml_iso88591_Entities':
62 if (count($this->xml_iso88591_Entities['in'])) {
65 for ($i = 0; $i < 32; $i++) {
66 $this->xml_iso88591_Entities["in"][] = chr($i);
67 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
70 for ($i = 160; $i < 256; $i++) {
71 $this->xml_iso88591_Entities["in"][] = chr($i);
72 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
75 /*case 'xml_cp1252_Entities':
76 if (count($this->xml_cp1252_Entities['in'])) {
79 for ($i = 128; $i < 160; $i++)
81 $this->xml_cp1252_Entities['in'][] = chr($i);
83 $this->xml_cp1252_Entities['out'] = array(
84 '€', '?', '‚', 'ƒ',
85 '„', '…', '†', '‡',
86 'ˆ', '‰', 'Š', '‹',
87 'Œ', '?', 'Ž', '?',
88 '?', '‘', '’', '“',
89 '”', '•', '–', '—',
90 '˜', '™', 'š', '›',
91 'œ', '?', 'ž', 'Ÿ'
93 $this->buildConversionTable('xml_iso88591_Entities');
96 throw new \Exception('Unsupported table: ' . $tableName);
101 * Convert a string to the correct XML representation in a target charset.
103 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
104 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
105 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
106 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
107 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
108 * bound by RFC 3023 to emit strict us-ascii.
110 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
111 * @todo make usage of iconv() or recode_string() or mb_string() where available
113 * @param string $data
114 * @param string $srcEncoding
115 * @param string $destEncoding
119 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
121 if ($srcEncoding == '') {
122 // lame, but we know no better...
123 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
126 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
127 switch ($conversion) {
129 case 'ISO-8859-1_US-ASCII':
130 $this->buildConversionTable('xml_iso88591_Entities');
131 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
132 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
135 case 'ISO-8859-1_UTF-8':
136 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
137 $escapedData = utf8_encode($escapedData);
140 case 'ISO-8859-1_ISO-8859-1':
141 case 'US-ASCII_US-ASCII':
142 case 'US-ASCII_UTF-8':
144 case 'US-ASCII_ISO-8859-1':
146 //case 'CP1252_CP1252':
147 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
151 case 'UTF-8_US-ASCII':
152 case 'UTF-8_ISO-8859-1':
153 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
155 // be kind to users creating string xmlrpc values out of different php types
156 $data = (string)$data;
158 for ($nn = 0; $nn < $ns; $nn++) {
161 // 7 bits: 0bbbbbbb (127)
163 /// @todo shall we replace this with a (supposedly) faster str_replace?
166 $escapedData .= '"';
169 $escapedData .= '&';
172 $escapedData .= ''';
175 $escapedData .= '<';
178 $escapedData .= '>';
183 } // 11 bits: 110bbbbb 10bbbbbb (2047)
184 elseif ($ii >> 5 == 6) {
186 $ii = ord($data[$nn + 1]);
188 $ii = ($b1 * 64) + $b2;
189 $ent = sprintf('&#%d;', $ii);
190 $escapedData .= $ent;
192 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
193 elseif ($ii >> 4 == 14) {
195 $ii = ord($data[$nn + 1]);
197 $ii = ord($data[$nn + 2]);
199 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
200 $ent = sprintf('&#%d;', $ii);
201 $escapedData .= $ent;
203 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
204 elseif ($ii >> 3 == 30) {
206 $ii = ord($data[$nn + 1]);
208 $ii = ord($data[$nn + 2]);
210 $ii = ord($data[$nn + 3]);
212 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
213 $ent = sprintf('&#%d;', $ii);
214 $escapedData .= $ent;
219 // when converting to latin-1, do not be so eager with using entities for characters 160-255
220 if ($conversion == 'UTF-8_ISO-8859-1') {
221 $this->buildConversionTable('xml_iso88591_Entities');
222 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
228 case 'CP1252_US-ASCII':
229 $this->buildConversionTable('xml_cp1252_Entities');
230 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
231 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
232 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
235 $this->buildConversionTable('xml_cp1252_Entities');
236 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
237 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all alone will NOT convert them)
238 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
239 $escapedData = utf8_encode($escapedData);
241 case 'CP1252_ISO-8859-1':
242 $this->buildConversionTable('xml_cp1252_Entities');
243 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
244 // 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...
245 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
251 Logger::instance()->errorLog('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
258 * Checks if a given charset encoding is present in a list of encodings or if it is a valid subset of any encoding
261 * @param string $encoding charset to be tested
262 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
266 public function isValidCharset($encoding, $validList)
268 if (is_string($validList)) {
269 $validList = explode(',', $validList);
271 if (@in_array(strtoupper($encoding), $validList)) {
274 if (array_key_exists($encoding, $this->charset_supersets)) {
275 foreach ($validList as $allowed) {
276 if (in_array($allowed, $this->charset_supersets[$encoding])) {
287 * Used only for backwards compatibility
290 * @param string $charset
294 * @throws \Exception for unknown/unsupported charsets
296 public function getEntities($charset)
298 //trigger_error('Method ' . __METHOD__ . ' is deprecated', E_USER_DEPRECATED);
303 return $this->xml_iso88591_Entities;
305 throw new \Exception('Unsupported charset: ' . $charset);