3 namespace PhpXmlRpc\Helper;
5 use PhpXmlRpc\PhpXmlRpc;
8 * @todo implement an interface
12 // tables used for transcoding different charsets into us-ascii xml
13 protected $xml_iso88591_Entities = array("in" => array(), "out" => array());
15 /// @todo should we add to the latin-1 table the characters from cp_1252 range, i.e. 128 to 159 ?
16 /// Those will NOT be present in true ISO-8859-1, but will save the unwary windows user from sending junk
17 /// (though no luck when receiving them...)
18 /// Note also that, apparently, while 'ISO/IEC 8859-1' has no characters defined for bytes 128 to 159,
19 /// IANA ISO-8859-1 does have well-defined 'C1' control codes for those - wikipedia's page on latin-1 says:
20 /// "ISO-8859-1 is the IANA preferred name for this standard when supplemented with the C0 and C1 control codes from ISO/IEC 6429."
21 /// Check what mbstring/iconv do by default with those?
23 //protected $xml_cp1252_Entities = array('in' => array(), out' => array());
25 protected $charset_supersets = array(
26 'US-ASCII' => array('ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4',
27 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8',
28 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-12',
29 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'UTF-8',
30 'EUC-JP', 'EUC-', 'EUC-KR', 'EUC-CN',),
33 /** @var Charset $instance */
34 protected static $instance = null;
37 * This class is singleton for performance reasons.
38 * @todo should we just make $xml_iso88591_Entities a static variable instead ?
42 public static function instance()
44 if (self::$instance === null) {
45 self::$instance = new static();
48 return self::$instance;
52 * Force usage as singleton
54 protected function __construct()
59 * @param string $tableName
60 * @throws \Exception for unsupported $tableName
62 protected function buildConversionTable($tableName)
65 case 'xml_iso88591_Entities':
66 if (count($this->xml_iso88591_Entities['in'])) {
69 for ($i = 0; $i < 32; $i++) {
70 $this->xml_iso88591_Entities["in"][] = chr($i);
71 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
74 for ($i = 160; $i < 256; $i++) {
75 $this->xml_iso88591_Entities["in"][] = chr($i);
76 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
79 /*case 'xml_cp1252_Entities':
80 if (count($this->xml_cp1252_Entities['in'])) {
83 for ($i = 128; $i < 160; $i++)
85 $this->xml_cp1252_Entities['in'][] = chr($i);
87 $this->xml_cp1252_Entities['out'] = array(
88 '€', '?', '‚', 'ƒ',
89 '„', '…', '†', '‡',
90 'ˆ', '‰', 'Š', '‹',
91 'Œ', '?', 'Ž', '?',
92 '?', '‘', '’', '“',
93 '”', '•', '–', '—',
94 '˜', '™', 'š', '›',
95 'œ', '?', 'ž', 'Ÿ'
97 $this->buildConversionTable('xml_iso88591_Entities');
100 throw new \Exception('Unsupported table: ' . $tableName);
105 * Convert a string to the correct XML representation in a target charset.
107 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
108 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
109 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
110 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
111 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
112 * bound by RFC 3023 to emit strict us-ascii.
114 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
115 * @todo make usage of iconv() or recode_string() or mb_string() where available
117 * @param string $data
118 * @param string $srcEncoding
119 * @param string $destEncoding
123 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
125 if ($srcEncoding == '') {
126 // lame, but we know no better...
127 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
130 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
131 switch ($conversion) {
133 case 'ISO-8859-1_US-ASCII':
134 $this->buildConversionTable('xml_iso88591_Entities');
135 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
136 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
139 case 'ISO-8859-1_UTF-8':
140 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
141 $escapedData = utf8_encode($escapedData);
144 case 'ISO-8859-1_ISO-8859-1':
145 case 'US-ASCII_US-ASCII':
146 case 'US-ASCII_UTF-8':
148 case 'US-ASCII_ISO-8859-1':
150 //case 'CP1252_CP1252':
151 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
155 case 'UTF-8_US-ASCII':
156 case 'UTF-8_ISO-8859-1':
157 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
159 // be kind to users creating string xmlrpc values out of different php types
160 $data = (string)$data;
162 for ($nn = 0; $nn < $ns; $nn++) {
165 // 7 bits: 0bbbbbbb (127)
167 /// @todo shall we replace this with a (supposedly) faster str_replace?
170 $escapedData .= '"';
173 $escapedData .= '&';
176 $escapedData .= ''';
179 $escapedData .= '<';
182 $escapedData .= '>';
187 } // 11 bits: 110bbbbb 10bbbbbb (2047)
188 elseif ($ii >> 5 == 6) {
190 $ii = ord($data[$nn + 1]);
192 $ii = ($b1 * 64) + $b2;
193 $ent = sprintf('&#%d;', $ii);
194 $escapedData .= $ent;
196 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
197 elseif ($ii >> 4 == 14) {
199 $ii = ord($data[$nn + 1]);
201 $ii = ord($data[$nn + 2]);
203 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
204 $ent = sprintf('&#%d;', $ii);
205 $escapedData .= $ent;
207 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
208 elseif ($ii >> 3 == 30) {
210 $ii = ord($data[$nn + 1]);
212 $ii = ord($data[$nn + 2]);
214 $ii = ord($data[$nn + 3]);
216 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
217 $ent = sprintf('&#%d;', $ii);
218 $escapedData .= $ent;
223 // when converting to latin-1, do not be so eager with using entities for characters 160-255
224 if ($conversion == 'UTF-8_ISO-8859-1') {
225 $this->buildConversionTable('xml_iso88591_Entities');
226 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
232 case 'CP1252_US-ASCII':
233 $this->buildConversionTable('xml_cp1252_Entities');
234 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
235 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
236 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
239 $this->buildConversionTable('xml_cp1252_Entities');
240 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
241 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all alone will NOT convert them)
242 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
243 $escapedData = utf8_encode($escapedData);
245 case 'CP1252_ISO-8859-1':
246 $this->buildConversionTable('xml_cp1252_Entities');
247 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
248 // 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...
249 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
255 Logger::instance()->errorLog('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
262 * Checks if a given charset encoding is present in a list of encodings or if it is a valid subset of any encoding
265 * @param string $encoding charset to be tested
266 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
270 public function isValidCharset($encoding, $validList)
272 if (is_string($validList)) {
273 $validList = explode(',', $validList);
275 if (@in_array(strtoupper($encoding), $validList)) {
278 if (array_key_exists($encoding, $this->charset_supersets)) {
279 foreach ($validList as $allowed) {
280 if (in_array($allowed, $this->charset_supersets[$encoding])) {
291 * Used only for backwards compatibility
294 * @param string $charset
298 * @throws \Exception for unknown/unsupported charsets
300 public function getEntities($charset)
302 //trigger_error('Method ' . __METHOD__ . ' is deprecated', E_USER_DEPRECATED);
307 return $this->xml_iso88591_Entities;
309 throw new \Exception('Unsupported charset: ' . $charset);