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(
21 '€', '?', '‚', 'ƒ',
22 '„', '…', '†', '‡',
23 'ˆ', '‰', 'Š', '‹',
24 'Œ', '?', 'Ž', '?',
25 '?', '‘', '’', '“',
26 '”', '•', '–', '—',
27 '˜', '™', 'š', '›',
28 'œ', '?', 'ž', 'Ÿ'
32 protected $charset_supersets = array(
33 'US-ASCII' => array('ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4',
34 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8',
35 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-12',
36 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'UTF-8',
37 'EUC-JP', 'EUC-', 'EUC-KR', 'EUC-CN',),
40 protected static $instance = null;
43 * This class is singleton for performance reasons.
44 * @todo can't we just make $xml_iso88591_Entities a static variable instead ?
48 public static function instance()
50 if (self::$instance === null) {
51 self::$instance = new self();
54 return self::$instance;
58 * @todo move the creation of the charset tables to be on-demand. This saves memory and time when latin-1 is not used at all
60 private function __construct()
62 for ($i = 0; $i < 32; $i++) {
63 $this->xml_iso88591_Entities["in"][] = chr($i);
64 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
67 for ($i = 160; $i < 256; $i++) {
68 $this->xml_iso88591_Entities["in"][] = chr($i);
69 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
72 /*for ($i = 128; $i < 160; $i++)
74 $this->xml_cp1252_Entities['in'][] = chr($i);
79 * Convert a string to the correct XML representation in a target charset.
81 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
82 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
83 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
84 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
85 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
86 * bound by RFC 3023 to emit strict us-ascii.
88 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
89 * @todo make usage of iconv() or recode_string() or mb_string() where available
92 * @param string $srcEncoding
93 * @param string $destEncoding
97 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
99 if ($srcEncoding == '') {
100 // lame, but we know no better...
101 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
104 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
105 switch ($conversion) {
107 case 'ISO-8859-1_US-ASCII':
108 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
109 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
112 case 'ISO-8859-1_UTF-8':
113 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
114 $escapedData = utf8_encode($escapedData);
117 case 'ISO-8859-1_ISO-8859-1':
118 case 'US-ASCII_US-ASCII':
119 case 'US-ASCII_UTF-8':
121 case 'US-ASCII_ISO-8859-1':
123 //case 'CP1252_CP1252':
124 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
128 case 'UTF-8_US-ASCII':
129 case 'UTF-8_ISO-8859-1':
130 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
132 // be kind to users creating string xmlrpc values out of different php types
133 $data = (string)$data;
135 for ($nn = 0; $nn < $ns; $nn++) {
138 // 7 bits: 0bbbbbbb (127)
140 /// @todo shall we replace this with a (supposedly) faster str_replace?
143 $escapedData .= '"';
146 $escapedData .= '&';
149 $escapedData .= ''';
152 $escapedData .= '<';
155 $escapedData .= '>';
160 } // 11 bits: 110bbbbb 10bbbbbb (2047)
161 elseif ($ii >> 5 == 6) {
163 $ii = ord($data[$nn + 1]);
165 $ii = ($b1 * 64) + $b2;
166 $ent = sprintf('&#%d;', $ii);
167 $escapedData .= $ent;
169 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
170 elseif ($ii >> 4 == 14) {
172 $ii = ord($data[$nn + 1]);
174 $ii = ord($data[$nn + 2]);
176 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
177 $ent = sprintf('&#%d;', $ii);
178 $escapedData .= $ent;
180 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
181 elseif ($ii >> 3 == 30) {
183 $ii = ord($data[$nn + 1]);
185 $ii = ord($data[$nn + 2]);
187 $ii = ord($data[$nn + 3]);
189 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
190 $ent = sprintf('&#%d;', $ii);
191 $escapedData .= $ent;
196 // when converting to latin-1, do not be so eager with using entities for characters 160-255
197 if ($conversion == 'UTF-8_ISO-8859-1') {
198 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
204 case 'CP1252_US-ASCII':
205 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
206 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
207 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
210 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
211 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all allone will NOT convert them)
212 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
213 $escapedData = utf8_encode($escapedData);
215 case 'CP1252_ISO-8859-1':
216 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
217 // 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...
218 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
224 Logger::instance()->errorLog('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
231 * Checks if a given charset encoding is present in a list of encodings or if it is a valid subset of any encoding
234 * @param string $encoding charset to be tested
235 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
239 public function isValidCharset($encoding, $validList)
241 if (is_string($validList)) {
242 $validList = explode(',', $validList);
244 if (@in_array(strtoupper($encoding), $validList)) {
247 if (array_key_exists($encoding, $this->charset_supersets)) {
248 foreach ($validList as $allowed) {
249 if (in_array($allowed, $this->charset_supersets[$encoding])) {
260 * Used only for backwards compatibility
263 * @param string $charset
267 * @throws \Exception for unknown/unsupported charsets
269 public function getEntities($charset)
271 //trigger_error('Method ' . __METHOD__ . ' is deprecated', E_USER_DEPRECATED);
276 return $this->xml_iso88591_Entities;
278 throw new \Exception('Unsupported charset: ' . $charset);