X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=src%2FEncoder.php;h=220ce88165a353f1f15faf99e3d39497e2ef9a19;hb=9d4cfc9508b03a579dd62a28c0e9e428b5052f9b;hp=8c639581b19a0f1c0d62cbd684ea7a4c599dde3f;hpb=8f01888ce85ef2e3748b236e55bbab62a363c2a1;p=plcapi.git
diff --git a/src/Encoder.php b/src/Encoder.php
index 8c63958..220ce88 100644
--- a/src/Encoder.php
+++ b/src/Encoder.php
@@ -4,6 +4,9 @@ namespace PhpXmlRpc;
use PhpXmlRpc\Helper\XMLParser;
+/**
+ * A helper class to easily convert between Value objects and php native values
+ */
class Encoder
{
/**
@@ -23,39 +26,39 @@ class Encoder
*
* @author Dan Libby (dan@libby.com)
*
- * @param Value|Request $xmlrpc_val
+ * @param Value|Request $xmlrpcVal
* @param array $options if 'decode_php_objs' is set in the options array, xmlrpc structs can be decoded into php objects; if 'dates_as_objects' is set xmlrpc datetimes are decoded as php DateTime objects (standard is
*
* @return mixed
*/
- public function decode($xmlrpc_val, $options = array())
+ public function decode($xmlrpcVal, $options = array())
{
- switch ($xmlrpc_val->kindOf()) {
+ switch ($xmlrpcVal->kindOf()) {
case 'scalar':
if (in_array('extension_api', $options)) {
- reset($xmlrpc_val->me);
- list($typ, $val) = each($xmlrpc_val->me);
+ reset($xmlrpcVal->me);
+ list($typ, $val) = each($xmlrpcVal->me);
switch ($typ) {
case 'dateTime.iso8601':
- $xmlrpc_val->scalar = $val;
- $xmlrpc_val->type = 'datetime';
- $xmlrpc_val->timestamp = \PhpXmlRpc\Helper\Date::iso8601_decode($val);
+ $xmlrpcVal->scalar = $val;
+ $xmlrpcVal->type = 'datetime';
+ $xmlrpcVal->timestamp = \PhpXmlRpc\Helper\Date::iso8601Decode($val);
- return $xmlrpc_val;
+ return $xmlrpcVal;
case 'base64':
- $xmlrpc_val->scalar = $val;
- $xmlrpc_val->type = $typ;
+ $xmlrpcVal->scalar = $val;
+ $xmlrpcVal->type = $typ;
- return $xmlrpc_val;
+ return $xmlrpcVal;
default:
- return $xmlrpc_val->scalarval();
+ return $xmlrpcVal->scalarval();
}
}
- if (in_array('dates_as_objects', $options) && $xmlrpc_val->scalartyp() == 'dateTime.iso8601') {
+ if (in_array('dates_as_objects', $options) && $xmlrpcVal->scalartyp() == 'dateTime.iso8601') {
// we return a Datetime object instead of a string
// since now the constructor of xmlrpc value accepts safely strings, ints and datetimes,
// we cater to all 3 cases here
- $out = $xmlrpc_val->scalarval();
+ $out = $xmlrpcVal->scalarval();
if (is_string($out)) {
$out = strtotime($out);
}
@@ -69,43 +72,41 @@ class Encoder
}
}
- return $xmlrpc_val->scalarval();
+ return $xmlrpcVal->scalarval();
case 'array':
- $size = $xmlrpc_val->arraysize();
$arr = array();
- for ($i = 0; $i < $size; $i++) {
- $arr[] = $this->decode($xmlrpc_val->arraymem($i), $options);
+ foreach($xmlrpcVal as $value) {
+ $arr[] = $this->decode($value, $options);
}
return $arr;
case 'struct':
- $xmlrpc_val->structreset();
// If user said so, try to rebuild php objects for specific struct vals.
/// @todo should we raise a warning for class not found?
// shall we check for proper subclass of xmlrpc value instead of
// presence of _php_class to detect what we can do?
- if (in_array('decode_php_objs', $options) && $xmlrpc_val->_php_class != ''
- && class_exists($xmlrpc_val->_php_class)
+ if (in_array('decode_php_objs', $options) && $xmlrpcVal->_php_class != ''
+ && class_exists($xmlrpcVal->_php_class)
) {
- $obj = @new $xmlrpc_val->_php_class();
- while (list($key, $value) = $xmlrpc_val->structeach()) {
+ $obj = @new $xmlrpcVal->_php_class();
+ foreach ($xmlrpcVal as $key => $value) {
$obj->$key = $this->decode($value, $options);
}
return $obj;
} else {
$arr = array();
- while (list($key, $value) = $xmlrpc_val->structeach()) {
+ foreach ($xmlrpcVal as $key => $value) {
$arr[$key] = $this->decode($value, $options);
}
return $arr;
}
case 'msg':
- $paramcount = $xmlrpc_val->getNumParams();
+ $paramCount = $xmlrpcVal->getNumParams();
$arr = array();
- for ($i = 0; $i < $paramcount; $i++) {
- $arr[] = $this->decode($xmlrpc_val->getParam($i));
+ for ($i = 0; $i < $paramCount; $i++) {
+ $arr[] = $this->decode($xmlrpcVal->getParam($i), $options);
}
return $arr;
@@ -125,44 +126,44 @@ class Encoder
*
* @author Dan Libby (dan@libby.com)
*
- * @param mixed $php_val the value to be converted into an xmlrpc value object
+ * @param mixed $phpVal the value to be converted into an xmlrpc value object
* @param array $options can include 'encode_php_objs', 'auto_dates', 'null_extension' or 'extension_api'
*
* @return \PhpXmlrpc\Value
*/
- public function encode($php_val, $options = array())
+ public function encode($phpVal, $options = array())
{
- $type = gettype($php_val);
+ $type = gettype($phpVal);
switch ($type) {
case 'string':
- if (in_array('auto_dates', $options) && preg_match('/^[0-9]{8}T[0-9]{2}:[0-9]{2}:[0-9]{2}$/', $php_val)) {
- $xmlrpc_val = new Value($php_val, Value::$xmlrpcDateTime);
+ if (in_array('auto_dates', $options) && preg_match('/^[0-9]{8}T[0-9]{2}:[0-9]{2}:[0-9]{2}$/', $phpVal)) {
+ $xmlrpcVal = new Value($phpVal, Value::$xmlrpcDateTime);
} else {
- $xmlrpc_val = new Value($php_val, Value::$xmlrpcString);
+ $xmlrpcVal = new Value($phpVal, Value::$xmlrpcString);
}
break;
case 'integer':
- $xmlrpc_val = new Value($php_val, Value::$xmlrpcInt);
+ $xmlrpcVal = new Value($phpVal, Value::$xmlrpcInt);
break;
case 'double':
- $xmlrpc_val = new Value($php_val, Value::$xmlrpcDouble);
+ $xmlrpcVal = new Value($phpVal, Value::$xmlrpcDouble);
break;
//
// Add support for encoding/decoding of booleans, since they are supported in PHP
case 'boolean':
- $xmlrpc_val = new Value($php_val, Value::$xmlrpcBoolean);
+ $xmlrpcVal = new Value($phpVal, Value::$xmlrpcBoolean);
break;
//
case 'array':
// PHP arrays can be encoded to either xmlrpc structs or arrays,
// depending on wheter they are hashes or plain 0..n integer indexed
// A shorter one-liner would be
- // $tmp = array_diff(array_keys($php_val), range(0, count($php_val)-1));
+ // $tmp = array_diff(array_keys($phpVal), range(0, count($phpVal)-1));
// but execution time skyrockets!
$j = 0;
$arr = array();
$ko = false;
- foreach ($php_val as $key => $val) {
+ foreach ($phpVal as $key => $val) {
$arr[$key] = $this->encode($val, $options);
if (!$ko && $key !== $j) {
$ko = true;
@@ -170,70 +171,90 @@ class Encoder
$j++;
}
if ($ko) {
- $xmlrpc_val = new Value($arr, Value::$xmlrpcStruct);
+ $xmlrpcVal = new Value($arr, Value::$xmlrpcStruct);
} else {
- $xmlrpc_val = new Value($arr, Value::$xmlrpcArray);
+ $xmlrpcVal = new Value($arr, Value::$xmlrpcArray);
}
break;
case 'object':
- if (is_a($php_val, 'PhpXmlRpc\Value')) {
- $xmlrpc_val = $php_val;
- } elseif (is_a($php_val, 'DateTime')) {
- $xmlrpc_val = new Value($php_val->format('Ymd\TH:i:s'), Value::$xmlrpcStruct);
+ if (is_a($phpVal, 'PhpXmlRpc\Value')) {
+ $xmlrpcVal = $phpVal;
+ } elseif (is_a($phpVal, 'DateTime')) {
+ $xmlrpcVal = new Value($phpVal->format('Ymd\TH:i:s'), Value::$xmlrpcStruct);
} else {
$arr = array();
- reset($php_val);
- while (list($k, $v) = each($php_val)) {
+ reset($phpVal);
+ while (list($k, $v) = each($phpVal)) {
$arr[$k] = $this->encode($v, $options);
}
- $xmlrpc_val = new Value($arr, Value::$xmlrpcStruct);
+ $xmlrpcVal = new Value($arr, Value::$xmlrpcStruct);
if (in_array('encode_php_objs', $options)) {
// let's save original class name into xmlrpc value:
// might be useful later on...
- $xmlrpc_val->_php_class = get_class($php_val);
+ $xmlrpcVal->_php_class = get_class($phpVal);
}
}
break;
case 'NULL':
if (in_array('extension_api', $options)) {
- $xmlrpc_val = new Value('', Value::$xmlrpcString);
+ $xmlrpcVal = new Value('', Value::$xmlrpcString);
} elseif (in_array('null_extension', $options)) {
- $xmlrpc_val = new Value('', Value::$xmlrpcNull);
+ $xmlrpcVal = new Value('', Value::$xmlrpcNull);
} else {
- $xmlrpc_val = new Value();
+ $xmlrpcVal = new Value();
}
break;
case 'resource':
if (in_array('extension_api', $options)) {
- $xmlrpc_val = new Value((int)$php_val, Value::$xmlrpcInt);
+ $xmlrpcVal = new Value((int)$phpVal, Value::$xmlrpcInt);
} else {
- $xmlrpc_val = new Value();
+ $xmlrpcVal = new Value();
}
// catch "user function", "unknown type"
default:
// giancarlo pinerolo
// it has to return
// an empty object in case, not a boolean.
- $xmlrpc_val = new Value();
+ $xmlrpcVal = new Value();
break;
}
- return $xmlrpc_val;
+ return $xmlrpcVal;
}
/**
* Convert the xml representation of a method response, method request or single
* xmlrpc value into the appropriate object (a.k.a. deserialize).
*
- * @param string $xml_val
+ * @param string $xmlVal
* @param array $options
*
* @return mixed false on error, or an instance of either Value, Request or Response
*/
- public function decode_xml($xml_val, $options = array())
+ public function decodeXml($xmlVal, $options = array())
{
+ // 'guestimate' encoding
+ $valEncoding = XMLParser::guessEncoding('', $xmlVal);
+ if ($valEncoding != '') {
+
+ // Since parsing will fail if charset is not specified in the xml prologue,
+ // the encoding is not UTF8 and there are non-ascii chars in the text, we try to work round that...
+ // The following code might be better for mb_string enabled installs, but
+ // makes the lib about 200% slower...
+ //if (!is_valid_charset($valEncoding, array('UTF-8'))
+ if (!in_array($valEncoding, array('UTF-8', 'US-ASCII')) && !XMLParser::hasEncoding($xmlVal)) {
+ if ($valEncoding == 'ISO-8859-1') {
+ $xmlVal = utf8_encode($xmlVal);
+ } else {
+ if (extension_loaded('mbstring')) {
+ $xmlVal = mb_convert_encoding($xmlVal, 'UTF-8', $valEncoding);
+ } else {
+ error_log('XML-RPC: ' . __METHOD__ . ': invalid charset encoding of xml text: ' . $valEncoding);
+ }
+ }
+ }
+ }
- /// @todo 'guestimate' encoding
$parser = xml_parser_create();
xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, true);
// What if internal encoding is not in one of the 3 allowed?
@@ -250,7 +271,7 @@ class Encoder
xml_set_element_handler($parser, 'xmlrpc_se_any', 'xmlrpc_ee');
xml_set_character_data_handler($parser, 'xmlrpc_cd');
xml_set_default_handler($parser, 'xmlrpc_dh');
- if (!xml_parse($parser, $xml_val, 1)) {
+ if (!xml_parse($parser, $xmlVal, 1)) {
$errstr = sprintf('XML error: %s at line %d, column %d',
xml_error_string(xml_get_error_code($parser)),
xml_get_current_line_number($parser), xml_get_current_column_number($parser));
@@ -271,8 +292,8 @@ class Encoder
case 'methodresponse':
$v = &$xmlRpcParser->_xh['value'];
if ($xmlRpcParser->_xh['isf'] == 1) {
- $vc = $v->structmem('faultCode');
- $vs = $v->structmem('faultString');
+ $vc = $v['faultCode'];
+ $vs = $v['faultString'];
$r = new Response(0, $vc->scalarval(), $vs->scalarval());
} else {
$r = new Response($v);
@@ -280,12 +301,12 @@ class Encoder
return $r;
case 'methodcall':
- $m = new Request($xmlRpcParser->_xh['method']);
+ $req = new Request($xmlRpcParser->_xh['method']);
for ($i = 0; $i < count($xmlRpcParser->_xh['params']); $i++) {
- $m->addParam($xmlRpcParser->_xh['params'][$i]);
+ $req->addParam($xmlRpcParser->_xh['params'][$i]);
}
- return $m;
+ return $req;
case 'value':
return $xmlRpcParser->_xh['value'];
default:
@@ -293,88 +314,4 @@ class Encoder
}
}
- /**
- * xml charset encoding guessing helper function.
- * Tries to determine the charset encoding of an XML chunk received over HTTP.
- * NB: according to the spec (RFC 3023), if text/xml content-type is received over HTTP without a content-type,
- * we SHOULD assume it is strictly US-ASCII. But we try to be more tolerant of unconforming (legacy?) clients/servers,
- * which will be most probably using UTF-8 anyway...
- *
- * @param string $httpheader the http Content-type header
- * @param string $xmlchunk xml content buffer
- * @param string $encoding_prefs comma separated list of character encodings to be used as default (when mb extension is enabled)
- * @return string
- *
- * @todo explore usage of mb_http_input(): does it detect http headers + post data? if so, use it instead of hand-detection!!!
- */
- public static function guess_encoding($httpheader = '', $xmlchunk = '', $encoding_prefs = null)
- {
- // discussion: see http://www.yale.edu/pclt/encoding/
- // 1 - test if encoding is specified in HTTP HEADERS
-
- //Details:
- // LWS: (\13\10)?( |\t)+
- // token: (any char but excluded stuff)+
- // quoted string: " (any char but double quotes and cointrol chars)* "
- // header: Content-type = ...; charset=value(; ...)*
- // where value is of type token, no LWS allowed between 'charset' and value
- // Note: we do not check for invalid chars in VALUE:
- // this had better be done using pure ereg as below
- // Note 2: we might be removing whitespace/tabs that ought to be left in if
- // the received charset is a quoted string. But nobody uses such charset names...
-
- /// @todo this test will pass if ANY header has charset specification, not only Content-Type. Fix it?
- $matches = array();
- if (preg_match('/;\s*charset\s*=([^;]+)/i', $httpheader, $matches)) {
- return strtoupper(trim($matches[1], " \t\""));
- }
-
- // 2 - scan the first bytes of the data for a UTF-16 (or other) BOM pattern
- // (source: http://www.w3.org/TR/2000/REC-xml-20001006)
- // NOTE: actually, according to the spec, even if we find the BOM and determine
- // an encoding, we should check if there is an encoding specified
- // in the xml declaration, and verify if they match.
- /// @todo implement check as described above?
- /// @todo implement check for first bytes of string even without a BOM? (It sure looks harder than for cases WITH a BOM)
- if (preg_match('/^(\x00\x00\xFE\xFF|\xFF\xFE\x00\x00|\x00\x00\xFF\xFE|\xFE\xFF\x00\x00)/', $xmlchunk)) {
- return 'UCS-4';
- } elseif (preg_match('/^(\xFE\xFF|\xFF\xFE)/', $xmlchunk)) {
- return 'UTF-16';
- } elseif (preg_match('/^(\xEF\xBB\xBF)/', $xmlchunk)) {
- return 'UTF-8';
- }
-
- // 3 - test if encoding is specified in the xml declaration
- // Details:
- // SPACE: (#x20 | #x9 | #xD | #xA)+ === [ \x9\xD\xA]+
- // EQ: SPACE?=SPACE? === [ \x9\xD\xA]*=[ \x9\xD\xA]*
- if (preg_match('/^<\?xml\s+version\s*=\s*' . "((?:\"[a-zA-Z0-9_.:-]+\")|(?:'[a-zA-Z0-9_.:-]+'))" .
- '\s+encoding\s*=\s*' . "((?:\"[A-Za-z][A-Za-z0-9._-]*\")|(?:'[A-Za-z][A-Za-z0-9._-]*'))/",
- $xmlchunk, $matches)) {
- return strtoupper(substr($matches[2], 1, -1));
- }
-
- // 4 - if mbstring is available, let it do the guesswork
- // NB: we favour finding an encoding that is compatible with what we can process
- if (extension_loaded('mbstring')) {
- if ($encoding_prefs) {
- $enc = mb_detect_encoding($xmlchunk, $encoding_prefs);
- } else {
- $enc = mb_detect_encoding($xmlchunk);
- }
- // NB: mb_detect likes to call it ascii, xml parser likes to call it US_ASCII...
- // IANA also likes better US-ASCII, so go with it
- if ($enc == 'ASCII') {
- $enc = 'US-' . $enc;
- }
-
- return $enc;
- } else {
- // no encoding specified: as per HTTP1.1 assume it is iso-8859-1?
- // Both RFC 2616 (HTTP 1.1) and 1945 (HTTP 1.0) clearly state that for text/xxx content types
- // this should be the standard. And we should be getting text/xml as request and response.
- // BUT we have to be backward compatible with the lib, which always used UTF-8 as default...
- return PhpXmlRpc::$xmlrpc_defencoding;
- }
- }
}