gggeek /
phpxmlrpc
Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like XMLParser often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use XMLParser, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 11 | class XMLParser |
||
| 12 | { |
||
| 13 | // used to store state during parsing |
||
| 14 | // quick explanation of components: |
||
| 15 | // ac - used to accumulate values |
||
| 16 | // stack - array with genealogy of xml elements names: |
||
| 17 | // used to validate nesting of xmlrpc elements |
||
| 18 | // valuestack - array used for parsing arrays and structs |
||
| 19 | // lv - used to indicate "looking for a value": implements |
||
| 20 | // the logic to allow values with no types to be strings |
||
| 21 | // isf - used to indicate a parsing fault (2) or xmlrpc response fault (1) |
||
| 22 | // isf_reason - used for storing xmlrpc response fault string |
||
| 23 | // method - used to store method name |
||
| 24 | // params - used to store parameters in method calls |
||
| 25 | // pt - used to store the type of each received parameter. Useful if parameters are automatically decoded to php values |
||
| 26 | // rt - 'methodcall or 'methodresponse' |
||
| 27 | public $_xh = array( |
||
| 28 | 'ac' => '', |
||
| 29 | 'stack' => array(), |
||
| 30 | 'valuestack' => array(), |
||
| 31 | 'isf' => 0, |
||
| 32 | 'isf_reason' => '', |
||
| 33 | 'method' => false, // so we can check later if we got a methodname or not |
||
| 34 | 'params' => array(), |
||
| 35 | 'pt' => array(), |
||
| 36 | 'rt' => '', |
||
| 37 | ); |
||
| 38 | |||
| 39 | public $xmlrpc_valid_parents = array( |
||
| 40 | 'VALUE' => array('MEMBER', 'DATA', 'PARAM', 'FAULT'), |
||
| 41 | 'BOOLEAN' => array('VALUE'), |
||
| 42 | 'I4' => array('VALUE'), |
||
| 43 | 'I8' => array('VALUE'), |
||
| 44 | 'EX:I8' => array('VALUE'), |
||
| 45 | 'INT' => array('VALUE'), |
||
| 46 | 'STRING' => array('VALUE'), |
||
| 47 | 'DOUBLE' => array('VALUE'), |
||
| 48 | 'DATETIME.ISO8601' => array('VALUE'), |
||
| 49 | 'BASE64' => array('VALUE'), |
||
| 50 | 'MEMBER' => array('STRUCT'), |
||
| 51 | 'NAME' => array('MEMBER'), |
||
| 52 | 'DATA' => array('ARRAY'), |
||
| 53 | 'ARRAY' => array('VALUE'), |
||
| 54 | 'STRUCT' => array('VALUE'), |
||
| 55 | 'PARAM' => array('PARAMS'), |
||
| 56 | 'METHODNAME' => array('METHODCALL'), |
||
| 57 | 'PARAMS' => array('METHODCALL', 'METHODRESPONSE'), |
||
| 58 | 'FAULT' => array('METHODRESPONSE'), |
||
| 59 | 'NIL' => array('VALUE'), // only used when extension activated |
||
| 60 | 'EX:NIL' => array('VALUE'), // only used when extension activated |
||
| 61 | ); |
||
| 62 | |||
| 63 | /** |
||
| 64 | * xml parser handler function for opening element tags. |
||
| 65 | */ |
||
| 66 | 16 | public function xmlrpc_se($parser, $name, $attrs, $acceptSingleVals = false) |
|
|
|
|||
| 67 | { |
||
| 68 | // if invalid xmlrpc already detected, skip all processing |
||
| 69 | 16 | if ($this->_xh['isf'] < 2) { |
|
| 70 | // check for correct element nesting |
||
| 71 | // top level element can only be of 2 types |
||
| 72 | /// @todo optimization creep: save this check into a bool variable, instead of using count() every time: |
||
| 73 | /// there is only a single top level element in xml anyway |
||
| 74 | 16 | if (count($this->_xh['stack']) == 0) { |
|
| 75 | 16 | if ($name != 'METHODRESPONSE' && $name != 'METHODCALL' && ( |
|
| 76 | 1 | $name != 'VALUE' && !$acceptSingleVals) |
|
| 77 | 16 | ) { |
|
| 78 | $this->_xh['isf'] = 2; |
||
| 79 | $this->_xh['isf_reason'] = 'missing top level xmlrpc element'; |
||
| 80 | |||
| 81 | return; |
||
| 82 | } else { |
||
| 83 | 16 | $this->_xh['rt'] = strtolower($name); |
|
| 84 | } |
||
| 85 | 16 | } else { |
|
| 86 | // not top level element: see if parent is OK |
||
| 87 | 16 | $parent = end($this->_xh['stack']); |
|
| 88 | 16 | if (!array_key_exists($name, $this->xmlrpc_valid_parents) || !in_array($parent, $this->xmlrpc_valid_parents[$name])) { |
|
| 89 | 2 | $this->_xh['isf'] = 2; |
|
| 90 | 2 | $this->_xh['isf_reason'] = "xmlrpc element $name cannot be child of $parent"; |
|
| 91 | |||
| 92 | 2 | return; |
|
| 93 | } |
||
| 94 | } |
||
| 95 | |||
| 96 | switch ($name) { |
||
| 97 | // optimize for speed switch cases: most common cases first |
||
| 98 | 16 | case 'VALUE': |
|
| 99 | /// @todo we could check for 2 VALUE elements inside a MEMBER or PARAM element |
||
| 100 | 14 | $this->_xh['vt'] = 'value'; // indicator: no value found yet |
|
| 101 | 14 | $this->_xh['ac'] = ''; |
|
| 102 | 14 | $this->_xh['lv'] = 1; |
|
| 103 | 14 | $this->_xh['php_class'] = null; |
|
| 104 | 14 | break; |
|
| 105 | 16 | case 'I8': |
|
| 106 | 16 | case 'EX:I8': |
|
| 107 | 1 | if (PHP_INT_SIZE === 4) { |
|
| 108 | /// INVALID ELEMENT: RAISE ISF so that it is later recognized!!! |
||
| 109 | $this->_xh['isf'] = 2; |
||
| 110 | $this->_xh['isf_reason'] = "Received i8 element but php is compiled in 32 bit mode"; |
||
| 111 | |||
| 112 | return; |
||
| 113 | } |
||
| 114 | // fall through voluntarily |
||
| 115 | 16 | case 'I4': |
|
| 116 | 16 | case 'INT': |
|
| 117 | 16 | case 'STRING': |
|
| 118 | 16 | case 'BOOLEAN': |
|
| 119 | 16 | case 'DOUBLE': |
|
| 120 | 16 | case 'DATETIME.ISO8601': |
|
| 121 | 16 | View Code Duplication | case 'BASE64': |
| 122 | 12 | if ($this->_xh['vt'] != 'value') { |
|
| 123 | // two data elements inside a value: an error occurred! |
||
| 124 | 1 | $this->_xh['isf'] = 2; |
|
| 125 | 1 | $this->_xh['isf_reason'] = "$name element following a {$this->_xh['vt']} element inside a single value"; |
|
| 126 | |||
| 127 | 1 | return; |
|
| 128 | } |
||
| 129 | 12 | $this->_xh['ac'] = ''; // reset the accumulator |
|
| 130 | 12 | break; |
|
| 131 | 16 | case 'STRUCT': |
|
| 132 | 16 | case 'ARRAY': |
|
| 133 | 13 | if ($this->_xh['vt'] != 'value') { |
|
| 134 | //two data elements inside a value: an error occurred! |
||
| 135 | 1 | $this->_xh['isf'] = 2; |
|
| 136 | 1 | $this->_xh['isf_reason'] = "$name element following a {$this->_xh['vt']} element inside a single value"; |
|
| 137 | |||
| 138 | 1 | return; |
|
| 139 | } |
||
| 140 | // create an empty array to hold child values, and push it onto appropriate stack |
||
| 141 | 12 | $curVal = array(); |
|
| 142 | 12 | $curVal['values'] = array(); |
|
| 143 | 12 | $curVal['type'] = $name; |
|
| 144 | // check for out-of-band information to rebuild php objs |
||
| 145 | // and in case it is found, save it |
||
| 146 | 12 | if (@isset($attrs['PHP_CLASS'])) { |
|
| 147 | $curVal['php_class'] = $attrs['PHP_CLASS']; |
||
| 148 | } |
||
| 149 | 12 | $this->_xh['valuestack'][] = $curVal; |
|
| 150 | 12 | $this->_xh['vt'] = 'data'; // be prepared for a data element next |
|
| 151 | 12 | break; |
|
| 152 | 16 | case 'DATA': |
|
| 153 | 2 | if ($this->_xh['vt'] != 'data') { |
|
| 154 | //two data elements inside a value: an error occurred! |
||
| 155 | 1 | $this->_xh['isf'] = 2; |
|
| 156 | 1 | $this->_xh['isf_reason'] = "found two data elements inside an array element"; |
|
| 157 | |||
| 158 | 1 | return; |
|
| 159 | } |
||
| 160 | 16 | case 'METHODCALL': |
|
| 161 | 16 | case 'METHODRESPONSE': |
|
| 162 | 16 | case 'PARAMS': |
|
| 163 | // valid elements that add little to processing |
||
| 164 | 16 | break; |
|
| 165 | 16 | case 'METHODNAME': |
|
| 166 | 16 | case 'NAME': |
|
| 167 | /// @todo we could check for 2 NAME elements inside a MEMBER element |
||
| 168 | 12 | $this->_xh['ac'] = ''; |
|
| 169 | 12 | break; |
|
| 170 | 16 | case 'FAULT': |
|
| 171 | 1 | $this->_xh['isf'] = 1; |
|
| 172 | 1 | break; |
|
| 173 | 16 | case 'MEMBER': |
|
| 174 | 11 | $this->_xh['valuestack'][count($this->_xh['valuestack']) - 1]['name'] = ''; // set member name to null, in case we do not find in the xml later on |
|
| 175 | //$this->_xh['ac']=''; |
||
| 176 | // Drop trough intentionally |
||
| 177 | 16 | case 'PARAM': |
|
| 178 | // clear value type, so we can check later if no value has been passed for this param/member |
||
| 179 | 16 | $this->_xh['vt'] = null; |
|
| 180 | 16 | break; |
|
| 181 | 1 | case 'NIL': |
|
| 182 | 1 | View Code Duplication | case 'EX:NIL': |
| 183 | 1 | if (PhpXmlRpc::$xmlrpc_null_extension) { |
|
| 184 | 1 | if ($this->_xh['vt'] != 'value') { |
|
| 185 | //two data elements inside a value: an error occurred! |
||
| 186 | $this->_xh['isf'] = 2; |
||
| 187 | $this->_xh['isf_reason'] = "$name element following a {$this->_xh['vt']} element inside a single value"; |
||
| 188 | |||
| 189 | return; |
||
| 190 | } |
||
| 191 | 1 | $this->_xh['ac'] = ''; // reset the accumulator |
|
| 192 | 1 | break; |
|
| 193 | } |
||
| 194 | // we do not support the <NIL/> extension, so |
||
| 195 | // drop through intentionally |
||
| 196 | 1 | default: |
|
| 197 | /// INVALID ELEMENT: RAISE ISF so that it is later recognized!!! |
||
| 198 | 1 | $this->_xh['isf'] = 2; |
|
| 199 | 1 | $this->_xh['isf_reason'] = "found not-xmlrpc xml element $name"; |
|
| 200 | 1 | break; |
|
| 201 | } |
||
| 202 | |||
| 203 | // Save current element name to stack, to validate nesting |
||
| 204 | 16 | $this->_xh['stack'][] = $name; |
|
| 205 | |||
| 206 | /// @todo optimization creep: move this inside the big switch() above |
||
| 207 | 16 | if ($name != 'VALUE') { |
|
| 208 | 16 | $this->_xh['lv'] = 0; |
|
| 209 | 16 | } |
|
| 210 | 16 | } |
|
| 211 | 16 | } |
|
| 212 | |||
| 213 | /** |
||
| 214 | * Used in decoding xml chunks that might represent single xmlrpc values. |
||
| 215 | */ |
||
| 216 | 3 | public function xmlrpc_se_any($parser, $name, $attrs) |
|
| 220 | |||
| 221 | /** |
||
| 222 | * xml parser handler function for close element tags. |
||
| 223 | */ |
||
| 224 | 16 | public function xmlrpc_ee($parser, $name, $rebuildXmlrpcvals = true) |
|
| 225 | { |
||
| 226 | 16 | if ($this->_xh['isf'] < 2) { |
|
| 227 | // push this element name from stack |
||
| 228 | // NB: if XML validates, correct opening/closing is guaranteed and |
||
| 229 | // we do not have to check for $name == $currElem. |
||
| 230 | // we also checked for proper nesting at start of elements... |
||
| 231 | 15 | $currElem = array_pop($this->_xh['stack']); |
|
| 232 | |||
| 233 | switch ($name) { |
||
| 234 | 15 | case 'VALUE': |
|
| 235 | // This if() detects if no scalar was inside <VALUE></VALUE> |
||
| 236 | 13 | if ($this->_xh['vt'] == 'value') { |
|
| 237 | 6 | $this->_xh['value'] = $this->_xh['ac']; |
|
| 238 | 6 | $this->_xh['vt'] = Value::$xmlrpcString; |
|
| 239 | 6 | } |
|
| 240 | |||
| 241 | 13 | if ($rebuildXmlrpcvals) { |
|
| 242 | // build the xmlrpc val out of the data received, and substitute it |
||
| 243 | 12 | $temp = new Value($this->_xh['value'], $this->_xh['vt']); |
|
| 244 | // in case we got info about underlying php class, save it |
||
| 245 | // in the object we're rebuilding |
||
| 246 | 12 | if (isset($this->_xh['php_class'])) { |
|
| 247 | $temp->_php_class = $this->_xh['php_class']; |
||
| 248 | } |
||
| 249 | // check if we are inside an array or struct: |
||
| 250 | // if value just built is inside an array, let's move it into array on the stack |
||
| 251 | 12 | $vscount = count($this->_xh['valuestack']); |
|
| 252 | 12 | View Code Duplication | if ($vscount && $this->_xh['valuestack'][$vscount - 1]['type'] == 'ARRAY') { |
| 253 | 2 | $this->_xh['valuestack'][$vscount - 1]['values'][] = $temp; |
|
| 254 | 2 | } else { |
|
| 255 | 11 | $this->_xh['value'] = $temp; |
|
| 256 | } |
||
| 257 | 12 | } else { |
|
| 258 | /// @todo this needs to treat correctly php-serialized objects, |
||
| 259 | /// since std deserializing is done by php_xmlrpc_decode, |
||
| 260 | /// which we will not be calling... |
||
| 261 | 3 | if (isset($this->_xh['php_class'])) { |
|
| 262 | } |
||
| 263 | |||
| 264 | // check if we are inside an array or struct: |
||
| 265 | // if value just built is inside an array, let's move it into array on the stack |
||
| 266 | 3 | $vscount = count($this->_xh['valuestack']); |
|
| 267 | 3 | View Code Duplication | if ($vscount && $this->_xh['valuestack'][$vscount - 1]['type'] == 'ARRAY') { |
| 268 | $this->_xh['valuestack'][$vscount - 1]['values'][] = $this->_xh['value']; |
||
| 269 | } |
||
| 270 | } |
||
| 271 | 13 | break; |
|
| 272 | 15 | case 'BOOLEAN': |
|
| 273 | 15 | case 'I4': |
|
| 274 | 15 | case 'I8': |
|
| 275 | 15 | case 'EX:I8': |
|
| 276 | 15 | case 'INT': |
|
| 277 | 15 | case 'STRING': |
|
| 278 | 15 | case 'DOUBLE': |
|
| 279 | 15 | case 'DATETIME.ISO8601': |
|
| 280 | 15 | case 'BASE64': |
|
| 281 | 12 | $this->_xh['vt'] = strtolower($name); |
|
| 282 | /// @todo: optimization creep - remove the if/elseif cycle below |
||
| 283 | /// since the case() in which we are already did that |
||
| 284 | 12 | if ($name == 'STRING') { |
|
| 285 | 5 | $this->_xh['value'] = $this->_xh['ac']; |
|
| 286 | 12 | View Code Duplication | } elseif ($name == 'DATETIME.ISO8601') { |
| 287 | 6 | if (!preg_match('/^[0-9]{8}T[0-9]{2}:[0-9]{2}:[0-9]{2}$/', $this->_xh['ac'])) { |
|
| 288 | error_log('XML-RPC: ' . __METHOD__ . ': invalid value received in DATETIME: ' . $this->_xh['ac']); |
||
| 289 | } |
||
| 290 | 6 | $this->_xh['vt'] = Value::$xmlrpcDateTime; |
|
| 291 | 6 | $this->_xh['value'] = $this->_xh['ac']; |
|
| 292 | 10 | } elseif ($name == 'BASE64') { |
|
| 293 | /// @todo check for failure of base64 decoding / catch warnings |
||
| 294 | $this->_xh['value'] = base64_decode($this->_xh['ac']); |
||
| 295 | 5 | } elseif ($name == 'BOOLEAN') { |
|
| 296 | // special case here: we translate boolean 1 or 0 into PHP |
||
| 297 | // constants true or false. |
||
| 298 | // Strings 'true' and 'false' are accepted, even though the |
||
| 299 | // spec never mentions them (see eg. Blogger api docs) |
||
| 300 | // NB: this simple checks helps a lot sanitizing input, ie no |
||
| 301 | // security problems around here |
||
| 302 | 2 | if ($this->_xh['ac'] == '1' || strcasecmp($this->_xh['ac'], 'true') == 0) { |
|
| 303 | 2 | $this->_xh['value'] = true; |
|
| 304 | 2 | } else { |
|
| 305 | // log if receiving something strange, even though we set the value to false anyway |
||
| 306 | 1 | if ($this->_xh['ac'] != '0' && strcasecmp($this->_xh['ac'], 'false') != 0) { |
|
| 307 | error_log('XML-RPC: ' . __METHOD__ . ': invalid value received in BOOLEAN: ' . $this->_xh['ac']); |
||
| 308 | } |
||
| 309 | 1 | $this->_xh['value'] = false; |
|
| 310 | } |
||
| 311 | 5 | View Code Duplication | } elseif ($name == 'DOUBLE') { |
| 312 | // we have a DOUBLE |
||
| 313 | // we must check that only 0123456789-.<space> are characters here |
||
| 314 | // NOTE: regexp could be much stricter than this... |
||
| 315 | 2 | if (!preg_match('/^[+-eE0123456789 \t.]+$/', $this->_xh['ac'])) { |
|
| 316 | /// @todo: find a better way of throwing an error than this! |
||
| 317 | error_log('XML-RPC: ' . __METHOD__ . ': non numeric value received in DOUBLE: ' . $this->_xh['ac']); |
||
| 318 | $this->_xh['value'] = 'ERROR_NON_NUMERIC_FOUND'; |
||
| 319 | } else { |
||
| 320 | // it's ok, add it on |
||
| 321 | 2 | $this->_xh['value'] = (double)$this->_xh['ac']; |
|
| 322 | } |
||
| 323 | 2 | } else { |
|
| 324 | // we have an I4/I8/INT |
||
| 325 | // we must check that only 0123456789-<space> are characters here |
||
| 326 | 4 | View Code Duplication | if (!preg_match('/^[+-]?[0123456789 \t]+$/', $this->_xh['ac'])) { |
| 327 | /// @todo find a better way of throwing an error than this! |
||
| 328 | error_log('XML-RPC: ' . __METHOD__ . ': non numeric value received in INT: ' . $this->_xh['ac']); |
||
| 329 | $this->_xh['value'] = 'ERROR_NON_NUMERIC_FOUND'; |
||
| 330 | } else { |
||
| 331 | // it's ok, add it on |
||
| 332 | 4 | $this->_xh['value'] = (int)$this->_xh['ac']; |
|
| 333 | } |
||
| 334 | } |
||
| 335 | 12 | $this->_xh['lv'] = 3; // indicate we've found a value |
|
| 336 | 12 | break; |
|
| 337 | 14 | case 'NAME': |
|
| 338 | 11 | $this->_xh['valuestack'][count($this->_xh['valuestack']) - 1]['name'] = $this->_xh['ac']; |
|
| 339 | 11 | break; |
|
| 340 | 14 | case 'MEMBER': |
|
| 341 | // add to array in the stack the last element built, |
||
| 342 | // unless no VALUE was found |
||
| 343 | 11 | if ($this->_xh['vt']) { |
|
| 344 | 11 | $vscount = count($this->_xh['valuestack']); |
|
| 345 | 11 | $this->_xh['valuestack'][$vscount - 1]['values'][$this->_xh['valuestack'][$vscount - 1]['name']] = $this->_xh['value']; |
|
| 346 | 11 | } else { |
|
| 347 | error_log('XML-RPC: ' . __METHOD__ . ': missing VALUE inside STRUCT in received xml'); |
||
| 348 | } |
||
| 349 | 11 | break; |
|
| 350 | 14 | case 'DATA': |
|
| 351 | 2 | $this->_xh['vt'] = null; // reset this to check for 2 data elements in a row - even if they're empty |
|
| 352 | 2 | break; |
|
| 353 | 13 | case 'STRUCT': |
|
| 354 | 13 | case 'ARRAY': |
|
| 355 | // fetch out of stack array of values, and promote it to current value |
||
| 356 | 11 | $currVal = array_pop($this->_xh['valuestack']); |
|
| 357 | 11 | $this->_xh['value'] = $currVal['values']; |
|
| 358 | 11 | $this->_xh['vt'] = strtolower($name); |
|
| 359 | 11 | if (isset($currVal['php_class'])) { |
|
| 360 | $this->_xh['php_class'] = $currVal['php_class']; |
||
| 361 | } |
||
| 362 | 11 | break; |
|
| 363 | 13 | case 'PARAM': |
|
| 364 | // add to array of params the current value, |
||
| 365 | // unless no VALUE was found |
||
| 366 | 11 | if ($this->_xh['vt']) { |
|
| 367 | 11 | $this->_xh['params'][] = $this->_xh['value']; |
|
| 368 | 11 | $this->_xh['pt'][] = $this->_xh['vt']; |
|
| 369 | 11 | } else { |
|
| 370 | error_log('XML-RPC: ' . __METHOD__ . ': missing VALUE inside PARAM in received xml'); |
||
| 371 | } |
||
| 372 | 11 | break; |
|
| 373 | 13 | case 'METHODNAME': |
|
| 374 | 2 | $this->_xh['method'] = preg_replace('/^[\n\r\t ]+/', '', $this->_xh['ac']); |
|
| 375 | 2 | break; |
|
| 376 | 12 | case 'NIL': |
|
| 377 | 12 | case 'EX:NIL': |
|
| 378 | 1 | if (PhpXmlRpc::$xmlrpc_null_extension) { |
|
| 379 | 1 | $this->_xh['vt'] = 'null'; |
|
| 380 | 1 | $this->_xh['value'] = null; |
|
| 381 | 1 | $this->_xh['lv'] = 3; |
|
| 382 | 1 | break; |
|
| 383 | } |
||
| 384 | // drop through intentionally if nil extension not enabled |
||
| 385 | 12 | case 'PARAMS': |
|
| 386 | 12 | case 'FAULT': |
|
| 387 | 12 | case 'METHODCALL': |
|
| 388 | 12 | case 'METHORESPONSE': |
|
| 389 | 12 | break; |
|
| 390 | 12 | default: |
|
| 391 | // End of INVALID ELEMENT! |
||
| 392 | // shall we add an assert here for unreachable code??? |
||
| 393 | 12 | break; |
|
| 394 | 12 | } |
|
| 395 | 15 | } |
|
| 396 | 16 | } |
|
| 397 | |||
| 398 | /** |
||
| 399 | * Used in decoding xmlrpc requests/responses without rebuilding xmlrpc Values. |
||
| 400 | */ |
||
| 401 | 3 | public function xmlrpc_ee_fast($parser, $name) |
|
| 405 | |||
| 406 | /** |
||
| 407 | * xml parser handler function for character data. |
||
| 408 | */ |
||
| 409 | 16 | public function xmlrpc_cd($parser, $data) |
|
| 420 | |||
| 421 | /** |
||
| 422 | * xml parser handler function for 'other stuff', ie. not char data or |
||
| 423 | * element start/end tag. In fact it only gets called on unknown entities... |
||
| 424 | */ |
||
| 425 | 3 | public function xmlrpc_dh($parser, $data) |
|
| 436 | |||
| 437 | /** |
||
| 438 | * xml charset encoding guessing helper function. |
||
| 439 | * Tries to determine the charset encoding of an XML chunk received over HTTP. |
||
| 440 | * NB: according to the spec (RFC 3023), if text/xml content-type is received over HTTP without a content-type, |
||
| 441 | * we SHOULD assume it is strictly US-ASCII. But we try to be more tolerant of non conforming (legacy?) clients/servers, |
||
| 442 | * which will be most probably using UTF-8 anyway... |
||
| 443 | * In order of importance checks: |
||
| 444 | * 1. http headers |
||
| 445 | * 2. BOM |
||
| 446 | * 3. XML declaration |
||
| 447 | * 4. guesses using mb_detect_encoding() |
||
| 448 | * |
||
| 449 | * @param string $httpHeader the http Content-type header |
||
| 450 | * @param string $xmlChunk xml content buffer |
||
| 451 | * @param string $encodingPrefs comma separated list of character encodings to be used as default (when mb extension is enabled). |
||
| 452 | * This can also be set globally using PhpXmlRpc::$xmlrpc_detectencodings |
||
| 453 | * @return string the encoding determined. Null if it can't be determined and mbstring is enabled, |
||
| 454 | * PhpXmlRpc::$xmlrpc_defencoding if it can't be determined and mbstring is not enabled |
||
| 455 | * |
||
| 456 | * @todo explore usage of mb_http_input(): does it detect http headers + post data? if so, use it instead of hand-detection!!! |
||
| 457 | */ |
||
| 458 | 16 | public static function guessEncoding($httpHeader = '', $xmlChunk = '', $encodingPrefs = null) |
|
| 459 | { |
||
| 460 | // discussion: see http://www.yale.edu/pclt/encoding/ |
||
| 461 | // 1 - test if encoding is specified in HTTP HEADERS |
||
| 462 | |||
| 463 | // Details: |
||
| 464 | // LWS: (\13\10)?( |\t)+ |
||
| 465 | // token: (any char but excluded stuff)+ |
||
| 466 | // quoted string: " (any char but double quotes and control chars)* " |
||
| 467 | // header: Content-type = ...; charset=value(; ...)* |
||
| 468 | // where value is of type token, no LWS allowed between 'charset' and value |
||
| 469 | // Note: we do not check for invalid chars in VALUE: |
||
| 470 | // this had better be done using pure ereg as below |
||
| 471 | // Note 2: we might be removing whitespace/tabs that ought to be left in if |
||
| 472 | // the received charset is a quoted string. But nobody uses such charset names... |
||
| 473 | |||
| 474 | /// @todo this test will pass if ANY header has charset specification, not only Content-Type. Fix it? |
||
| 475 | 16 | $matches = array(); |
|
| 476 | 16 | if (preg_match('/;\s*charset\s*=([^;]+)/i', $httpHeader, $matches)) { |
|
| 477 | 2 | return strtoupper(trim($matches[1], " \t\"")); |
|
| 478 | } |
||
| 479 | |||
| 480 | // 2 - scan the first bytes of the data for a UTF-16 (or other) BOM pattern |
||
| 481 | // (source: http://www.w3.org/TR/2000/REC-xml-20001006) |
||
| 482 | // NOTE: actually, according to the spec, even if we find the BOM and determine |
||
| 483 | // an encoding, we should check if there is an encoding specified |
||
| 484 | // in the xml declaration, and verify if they match. |
||
| 485 | /// @todo implement check as described above? |
||
| 486 | /// @todo implement check for first bytes of string even without a BOM? (It sure looks harder than for cases WITH a BOM) |
||
| 487 | 16 | if (preg_match('/^(\x00\x00\xFE\xFF|\xFF\xFE\x00\x00|\x00\x00\xFF\xFE|\xFE\xFF\x00\x00)/', $xmlChunk)) { |
|
| 488 | return 'UCS-4'; |
||
| 489 | 16 | } elseif (preg_match('/^(\xFE\xFF|\xFF\xFE)/', $xmlChunk)) { |
|
| 490 | return 'UTF-16'; |
||
| 491 | 16 | } elseif (preg_match('/^(\xEF\xBB\xBF)/', $xmlChunk)) { |
|
| 492 | return 'UTF-8'; |
||
| 493 | } |
||
| 494 | |||
| 495 | // 3 - test if encoding is specified in the xml declaration |
||
| 496 | // Details: |
||
| 497 | // SPACE: (#x20 | #x9 | #xD | #xA)+ === [ \x9\xD\xA]+ |
||
| 498 | // EQ: SPACE?=SPACE? === [ \x9\xD\xA]*=[ \x9\xD\xA]* |
||
| 499 | 16 | if (preg_match('/^<\?xml\s+version\s*=\s*' . "((?:\"[a-zA-Z0-9_.:-]+\")|(?:'[a-zA-Z0-9_.:-]+'))" . |
|
| 500 | 16 | '\s+encoding\s*=\s*' . "((?:\"[A-Za-z][A-Za-z0-9._-]*\")|(?:'[A-Za-z][A-Za-z0-9._-]*'))/", |
|
| 501 | 16 | $xmlChunk, $matches)) { |
|
| 502 | 2 | return strtoupper(substr($matches[2], 1, -1)); |
|
| 503 | } |
||
| 504 | |||
| 505 | // 4 - if mbstring is available, let it do the guesswork |
||
| 506 | 14 | if (extension_loaded('mbstring')) { |
|
| 507 | 14 | if ($encodingPrefs == null && PhpXmlRpc::$xmlrpc_detectencodings != null) { |
|
| 508 | $encodingPrefs = PhpXmlRpc::$xmlrpc_detectencodings; |
||
| 509 | } |
||
| 510 | 14 | if ($encodingPrefs) { |
|
| 511 | $enc = mb_detect_encoding($xmlChunk, $encodingPrefs); |
||
| 512 | } else { |
||
| 513 | 14 | $enc = mb_detect_encoding($xmlChunk); |
|
| 514 | } |
||
| 515 | // NB: mb_detect likes to call it ascii, xml parser likes to call it US_ASCII... |
||
| 516 | // IANA also likes better US-ASCII, so go with it |
||
| 517 | 14 | if ($enc == 'ASCII') { |
|
| 518 | 13 | $enc = 'US-' . $enc; |
|
| 519 | 13 | } |
|
| 520 | |||
| 521 | 14 | return $enc; |
|
| 522 | } else { |
||
| 523 | // no encoding specified: as per HTTP1.1 assume it is iso-8859-1? |
||
| 524 | // Both RFC 2616 (HTTP 1.1) and 1945 (HTTP 1.0) clearly state that for text/xxx content types |
||
| 525 | // this should be the standard. And we should be getting text/xml as request and response. |
||
| 526 | // BUT we have to be backward compatible with the lib, which always used UTF-8 as default... |
||
| 527 | return PhpXmlRpc::$xmlrpc_defencoding; |
||
| 528 | } |
||
| 529 | } |
||
| 530 | |||
| 531 | /** |
||
| 532 | * Helper function: checks if an xml chunk as a charset declaration (BOM or in the xml declaration) |
||
| 533 | * |
||
| 534 | * @param string $xmlChunk |
||
| 535 | * @return bool |
||
| 536 | */ |
||
| 537 | 1 | public static function hasEncoding($xmlChunk) |
|
| 561 | } |
||
| 562 |
Loading history...
This check looks from parameters that have been defined for a function or method, but which are not used in the method body.