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 /** @var Charset $instance */
31 protected static $instance = null;
34 * This class is singleton for performance reasons.
35 * @todo can't we just make $xml_iso88591_Entities a static variable instead ?
39 public static function instance()
41 if (self::$instance === null) {
42 self::$instance = new static();
45 return self::$instance;
49 * Force usage as singleton
51 protected function __construct()
56 * @param string $tableName
57 * @throws \Exception for unsupported $tableName
59 protected function buildConversionTable($tableName)
62 case 'xml_iso88591_Entities':
63 if (count($this->xml_iso88591_Entities['in'])) {
66 for ($i = 0; $i < 32; $i++) {
67 $this->xml_iso88591_Entities["in"][] = chr($i);
68 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
71 for ($i = 160; $i < 256; $i++) {
72 $this->xml_iso88591_Entities["in"][] = chr($i);
73 $this->xml_iso88591_Entities["out"][] = "&#{$i};";
76 /*case 'xml_cp1252_Entities':
77 if (count($this->xml_cp1252_Entities['in'])) {
80 for ($i = 128; $i < 160; $i++)
82 $this->xml_cp1252_Entities['in'][] = chr($i);
84 $this->xml_cp1252_Entities['out'] = array(
85 '€', '?', '‚', 'ƒ',
86 '„', '…', '†', '‡',
87 'ˆ', '‰', 'Š', '‹',
88 'Œ', '?', 'Ž', '?',
89 '?', '‘', '’', '“',
90 '”', '•', '–', '—',
91 '˜', '™', 'š', '›',
92 'œ', '?', 'ž', 'Ÿ'
94 $this->buildConversionTable('xml_iso88591_Entities');
97 throw new \Exception('Unsupported table: ' . $tableName);
102 * Convert a string to the correct XML representation in a target charset.
104 * To help correct communication of non-ascii chars inside strings, regardless of the charset used when sending
105 * requests, parsing them, sending responses and parsing responses, an option is to convert all non-ascii chars
106 * present in the message into their equivalent 'charset entity'. Charset entities enumerated this way are
107 * independent of the charset encoding used to transmit them, and all XML parsers are bound to understand them.
108 * Note that in the std case we are not sending a charset encoding mime type along with http headers, so we are
109 * bound by RFC 3023 to emit strict us-ascii.
111 * @todo do a bit of basic benchmarking (strtr vs. str_replace)
112 * @todo make usage of iconv() or recode_string() or mb_string() where available
114 * @param string $data
115 * @param string $srcEncoding
116 * @param string $destEncoding
120 public function encodeEntities($data, $srcEncoding = '', $destEncoding = '')
122 if ($srcEncoding == '') {
123 // lame, but we know no better...
124 $srcEncoding = PhpXmlRpc::$xmlrpc_internalencoding;
127 $conversion = strtoupper($srcEncoding . '_' . $destEncoding);
128 switch ($conversion) {
130 case 'ISO-8859-1_US-ASCII':
131 $this->buildConversionTable('xml_iso88591_Entities');
132 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
133 $escapedData = str_replace($this->xml_iso88591_Entities['in'], $this->xml_iso88591_Entities['out'], $escapedData);
136 case 'ISO-8859-1_UTF-8':
137 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
138 $escapedData = utf8_encode($escapedData);
141 case 'ISO-8859-1_ISO-8859-1':
142 case 'US-ASCII_US-ASCII':
143 case 'US-ASCII_UTF-8':
145 case 'US-ASCII_ISO-8859-1':
147 //case 'CP1252_CP1252':
148 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
152 case 'UTF-8_US-ASCII':
153 case 'UTF-8_ISO-8859-1':
154 // NB: this will choke on invalid UTF-8, going most likely beyond EOF
156 // be kind to users creating string xmlrpc values out of different php types
157 $data = (string)$data;
159 for ($nn = 0; $nn < $ns; $nn++) {
162 // 7 bits: 0bbbbbbb (127)
164 /// @todo shall we replace this with a (supposedly) faster str_replace?
167 $escapedData .= '"';
170 $escapedData .= '&';
173 $escapedData .= ''';
176 $escapedData .= '<';
179 $escapedData .= '>';
184 } // 11 bits: 110bbbbb 10bbbbbb (2047)
185 elseif ($ii >> 5 == 6) {
187 $ii = ord($data[$nn + 1]);
189 $ii = ($b1 * 64) + $b2;
190 $ent = sprintf('&#%d;', $ii);
191 $escapedData .= $ent;
193 } // 16 bits: 1110bbbb 10bbbbbb 10bbbbbb
194 elseif ($ii >> 4 == 14) {
196 $ii = ord($data[$nn + 1]);
198 $ii = ord($data[$nn + 2]);
200 $ii = ((($b1 * 64) + $b2) * 64) + $b3;
201 $ent = sprintf('&#%d;', $ii);
202 $escapedData .= $ent;
204 } // 21 bits: 11110bbb 10bbbbbb 10bbbbbb 10bbbbbb
205 elseif ($ii >> 3 == 30) {
207 $ii = ord($data[$nn + 1]);
209 $ii = ord($data[$nn + 2]);
211 $ii = ord($data[$nn + 3]);
213 $ii = ((((($b1 * 64) + $b2) * 64) + $b3) * 64) + $b4;
214 $ent = sprintf('&#%d;', $ii);
215 $escapedData .= $ent;
220 // when converting to latin-1, do not be so eager with using entities for characters 160-255
221 if ($conversion == 'UTF-8_ISO-8859-1') {
222 $this->buildConversionTable('xml_iso88591_Entities');
223 $escapedData = str_replace(array_slice($this->xml_iso88591_Entities['out'], 32), array_slice($this->xml_iso88591_Entities['in'], 32), $escapedData);
229 case 'CP1252_US-ASCII':
230 $this->buildConversionTable('xml_cp1252_Entities');
231 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
232 $escapedData = str_replace($this->xml_iso88591_Entities']['in'], $this->xml_iso88591_Entities['out'], $escapedData);
233 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
236 $this->buildConversionTable('xml_cp1252_Entities');
237 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
238 /// @todo we could use real UTF8 chars here instead of xml entities... (note that utf_8 encode all alone will NOT convert them)
239 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
240 $escapedData = utf8_encode($escapedData);
242 case 'CP1252_ISO-8859-1':
243 $this->buildConversionTable('xml_cp1252_Entities');
244 $escapedData = str_replace(array('&', '"', "'", '<', '>'), array('&', '"', ''', '<', '>'), $data);
245 // 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...
246 $escapedData = str_replace($this->xml_cp1252_Entities['in'], $this->xml_cp1252_Entities['out'], $escapedData);
252 Logger::instance()->errorLog('XML-RPC: ' . __METHOD__ . ": Converting from $srcEncoding to $destEncoding: not supported...");
259 * Checks if a given charset encoding is present in a list of encodings or if it is a valid subset of any encoding
262 * @param string $encoding charset to be tested
263 * @param string|array $validList comma separated list of valid charsets (or array of charsets)
267 public function isValidCharset($encoding, $validList)
269 if (is_string($validList)) {
270 $validList = explode(',', $validList);
272 if (@in_array(strtoupper($encoding), $validList)) {
275 if (array_key_exists($encoding, $this->charset_supersets)) {
276 foreach ($validList as $allowed) {
277 if (in_array($allowed, $this->charset_supersets[$encoding])) {
288 * Used only for backwards compatibility
291 * @param string $charset
295 * @throws \Exception for unknown/unsupported charsets
297 public function getEntities($charset)
299 //trigger_error('Method ' . __METHOD__ . ' is deprecated', E_USER_DEPRECATED);
304 return $this->xml_iso88591_Entities;
306 throw new \Exception('Unsupported charset: ' . $charset);