diff options
| -rw-r--r-- | it_xml.class | 11 | ||||
| -rwxr-xr-x | tests/it_xml.t | 35 |
2 files changed, 26 insertions, 20 deletions
diff --git a/it_xml.class b/it_xml.class index 0679c69..f854682 100644 --- a/it_xml.class +++ b/it_xml.class @@ -29,7 +29,7 @@ class it_xml * @param $p associative array * @param $p['forcearray'] xml tags to ALWAYS return as array * @param $p['safety'] 2 causes program abort with invalid xml, 1 (default) causes error report, 0 just returns false - * @param $p['encoding'] Output character encoding (e.g. UTF-8, default: ISO-8859-1) + * @param $p['encoding'] Output character encoding (utf-8, iso-8859-1 or us-ascii, default: ini_get('default_charset') * @param $p['prefix'] Optional prefix for class names * @param $p['lowercase'] Lowercase all tag and attribute names * @return XML object tree or null on failure @@ -49,23 +49,22 @@ function it_xml($xmldata = "", $p = array()) function create($xmldata, $p = array()) { $xml = new it_xml; - return $xml->from_xml($xmldata, array('factory' => true) + $p) ? $xml->_root : null; } function from_xml($xmldata, $p) { - $this->_p = $p + array('encoding' => "ISO-8859-1", 'safety' => 1); + $this->_p = $p + array('encoding' => ini_get('default_charset'), 'safety' => 1); $this->_arrayforce = array_flip((array)$this->_p['forcearray']); $this->_stack = array(); unset($this->error); - $parser = xml_parser_create($this->_p['encoding']); + $parser = xml_parser_create(); xml_set_object($parser, $this); xml_set_element_handler($parser, "start_element", "end_element"); xml_set_character_data_handler($parser, "character_data"); xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, 0); xml_parser_set_option($parser, XML_OPTION_TARGET_ENCODING, $this->_p['encoding']); - + $result = true; if (is_resource($xmldata)) @@ -123,7 +122,7 @@ function _sanitize($xmldata, $isutf8 = null) if (!preg_match('/^(<\?xml|\xEF\xBB\xBF|\xFE\xFF|\xFF\xFE|\x00\x00\xFE\xFF|\x00\x00\xFF\xFE)/', $xmldata)) $xmldata = '<?xml version="1.0" encoding="' . $this->_p['encoding'] . '"?>' . $xmldata; - $isutf8 = (!preg_match('/^<\?xml[^>]* encoding=/i', $xmldata) || preg_match('/^<\?xml[^>]* encoding=.UTF-8/i', $xmldata)); + $isutf8 = (!preg_match('/^<\?xml[^>]* encoding=/i', $xmldata) || preg_match('/^<\?xml[^>]* encoding=.utf-8/i', $xmldata)); } # Decode illegal entities but protect semantically important ones diff --git a/tests/it_xml.t b/tests/it_xml.t index f74c54b..d74fadf 100755 --- a/tests/it_xml.t +++ b/tests/it_xml.t @@ -5,13 +5,14 @@ function match($xmldata, $expected, $name, $prefix = "", $p = array()) { - $classname = $prefix ? ($prefix . "_xml") : "it_xml"; + $classname = ($prefix ?: "it") . "_xml"; $varname = $prefix . "foo"; $xmldata = "<root>$xmldata</root>"; $xml = new $classname($xmldata, $p); + $mod_utf8 = $p['encoding'] != "iso-8859-1" ? "u" : ""; is( - preg_replace('/[#\s]+/', " ", print_r($xml->$varname, true)), + preg_replace('/[#\s]+/' . $mod_utf8, " ", print_r($xml->$varname, true)), $expected, "$name (string)" ); @@ -24,11 +25,10 @@ function match($xmldata, $expected, $name, $prefix = "", $p = array()) fclose($tmpfile); is( - preg_replace('/[#\s]+/', " ", print_r($xml->$varname, true)), + preg_replace('/[#\s]+/' . $mod_utf8, " ", print_r($xml->$varname, true)), $expected, "$name (file)" ); - } match( @@ -44,8 +44,8 @@ match( ); match( - '<foo title="Zürich">Stüssihofstadt</foo>', - 'foo Object ( [attr] => Array ( [title] => Zürich ) [val] => Stüssihofstadt ) ', + '<foo title="Zürich">Stüssihofstadt</foo>', + 'foo Object ( [attr] => Array ( [title] => Zürich ) [val] => Stüssihofstadt ) ', 'simple tag with latin1 content and attribute' ); @@ -62,26 +62,33 @@ match( ); match( - '<foo>&amp; <a> &amp; <b> &amp; <c> ü</foo>', - 'foo Object ( [val] => & <a> & <b> & <c> ü ) ', - 'Predecode illegal entities while keeping properly encoded ones' + '<foo>x ü y</foo>', + utf8_decode('foo Object ( [val] => x ü y ) '), + 'Manual encoding override', + "", + array('encoding' => "iso-8859-1") ); match( '<foo>&amp; <a> &amp; <b> &amp; <c> ü</foo>', - utf8_encode('foo Object ( [val] => & <a> & <b> & <c> ü ) '), - 'Predecode illegal entities while keeping properly encoded ones (UTF-8)', - "", - array('encoding' => "UTF-8") + 'foo Object ( [val] => & <a> & <b> & <c> ü ) ', + 'Predecode illegal entities while keeping properly encoded ones', ); +match( + '<foo>&amp; <a> &amp; <b> &amp; <c> ü</foo>', + utf8_decode('foo Object ( [val] => & <a> & <b> & <c> ü ) '), + 'Predecode illegal entities while keeping properly encoded ones (iso-8859-1)', + "", + array('encoding' => "iso-8859-1") +); match( "<foo>a\x05b</foo>", 'foo Object ( [val] => a b ) ', 'Illegal latin 1 character', "", - array('encoding' => "ISO-8859-1") + array('encoding' => "iso-8859-1") ); # Test inheritance |