PHP4环境下分析解析XML

　

  转载 http://blog.s135.org

今天在PHP4环境下重新写一个接口程序，需要大量分析解析XML，PHP的xml_parse_into_struct()函数不能直接生成便于使用的数组，而SimpleXML扩展在PHP5中才支持，于是逛逛搜索引擎，在老外的网站上找到了一个不错的PHP XML操作类。

一、用法举例：
1、将XML文件解释成便于使用的数组：

include('xml.php');    //引用PHP XML操作类
$xml = file_get_contents('data.xml');    //读取XML文件
//$xml = file_get_contents("php://input");    //读取POST过来的输入流
$data=XML_unserialize($xml);
echo '';
print_r($data);
echo '';
?>

<?php include('xml.php');    //引用PHP XML操作类 $xml = file_get_contents('data.xml');    //读取XML文件 //$xml = file_get_contents("php://input");    //读取POST过来的输入流 $data=XML_unserialize($xml); echo '<pre>'; print_r($data); echo '</pre>'; ?>

data.xml文件：

xml version="1.0" encoding="GBK"?>
<video>
<upload>
<videoid>998videoid>
<name>name>
<memo>memo>
<up_userid>11317up_userid>
upload>
video>

<?xml version="1.0" encoding="GBK"?> <video> <upload> <videoid>998</videoid> <name><![CDATA[回忆未来]]></name> <memo><![CDATA[def]]></memo> <up_userid>11317</up_userid> </upload> </video>

利用该XML操作类生成的对应数组（汉字编码：UTF-8）：

Array
(
    [video] => Array
        (
            [upload] => Array
                (
                    [videoid] => 998
                    [name] => 回忆未来
                    [memo] => def
                    [up_userid] => 11317
                )

        )

)

Array (    [video] => Array        (            [upload] => Array                (                    [videoid] => 998                    [name] => 回忆未来                    [memo] => def                    [up_userid] => 11317                )        ) )

2、将数组转换成XML文件：

include('xml.php');//引用PHP XML操作类
$xml = XML_serialize($data);
?>

<?php include('xml.php');//引用PHP XML操作类 $xml = XML_serialize($data); ?>

二、PHP XML操作类源代码：

###################################################################################
#
# XML Library, by Keith Devens, version 1.2b
# "http://keithdevens.com/software/phpxml" target="_blank">http://keithdevens.com/software/phpxml
#
# This code is Open Source, released under terms similar to the Artistic License.
# Read the license at "http://keithdevens.com/software/license" target="_blank">http://keithdevens.com/software/license
#
###################################################################################

###################################################################################
# XML_unserialize: takes raw XML as a parameter (a string)
# and returns an equivalent PHP data structure
###################################################################################
function & XML_unserialize(&$xml){
    $xml_parser = &new XML();
    $data = &$xml_parser->parse($xml);
    $xml_parser->destruct();
    return $data;
}
###################################################################################
# XML_serialize: serializes any PHP data structure into XML
# Takes one parameter: the data to serialize. Must be an array.
###################################################################################
function & XML_serialize(&$data, $level = 0, $prior_key = NULL){
    if($level == 0){ ob_start(); echo '',"\n"; }
    while(list($key, $value) = each($data))
        if(!strpos($key, ' attr')) #if it's not an attribute
            #we don't treat attributes by themselves, so for an emptyempty element
            # that has attributes you still need to set the element to NULL

            if(is_array($value) and array_key_exists(0, $value)){
                XML_serialize($value, $level, $key);
            }else{
                $tag = $prior_key ? $prior_key : $key;
                echo str_repeat("\t", $level),'<',$tag;
                if(array_key_exists("$key attr", $data)){ #if there's an attribute for this element
                    while(list($attr_name, $attr_value) = each($data["$key attr"]))
                        echo ' ',$attr_name,'="',htmlspecialchars($attr_value),'"';
                    reset($data["$key attr"]);
                }

                if(is_null($value)) echo " />\n";
                elseif(!is_array($value)) echo '>',htmlspecialchars($value),"\n";
                else echo ">\n",XML_serialize($value, $level+1),str_repeat("\t", $level),"\n";
            }
    reset($data);
    if($level == 0){ $str = &ob_get_contents(); ob_end_clean(); return $str; }
}
###################################################################################
# XML class: utility class to be used with PHP's XML handling functions
###################################################################################
class XML{
    var $parser;   #a reference to the XML parser
    var $document; #the entire XML structure built up so far
    var $parent;   #a pointer to the current parent - the parent will be an array
    var $stack;    #a stack of the most recent parent at each nesting level
    var $last_opened_tag; #keeps track of the last tag opened.

    function XML(){
         $this->parser = &xml_parser_create();
        xml_parser_set_option(&$this->parser, XML_OPTION_CASE_FOLDING, false);
        xml_set_object(&$this->parser, &$this);
        xml_set_element_handler(&$this->parser, 'open','close');
        xml_set_character_data_handler(&$this->parser, 'data');
    }
    function destruct(){ xml_parser_free(&$this->parser); }
    function & parse(&$data){
        $this->document = array();
        $this->stack    = array();
        $this->parent   = &$this->document;
        return xml_parse(&$this->parser, &$data, true) ? $this->document : NULL;
    }
    function open(&$parser, $tag, $attributes){
        $this->data = ''; #stores temporary cdata
        $this->last_opened_tag = $tag;
        if(is_array($this->parent) and array_key_exists($tag,$this->parent)){ #if you've seen this tag before
            if(is_array($this->parent[$tag]) and array_key_exists(0,$this->parent[$tag])){ #if the keys are numeric
                #this is the third or later instance of $tag we've come across
                $key = count_numeric_items($this->parent[$tag]);
            }else{
                #this is the second instance of $tag that we've seen. shift around
                if(array_key_exists("$tag attr",$this->parent)){
                    $arr = array('0 attr'=>&$this->parent["$tag attr"], &$this->parent[$tag]);
                    unset($this->parent["$tag attr"]);
                }else{
                    $arr = array(&$this->parent[$tag]);
                }
                $this->parent[$tag] = &$arr;
                $key = 1;
            }
            $this->parent = &$this->parent[$tag];
        }else{
            $key = $tag;
        }
        if($attributes) $this->parent["$key attr"] = $attributes;
        $this->parent  = &$this->parent[$key];
        $this->stack[] = &$this->parent;
    }
    function data(&$parser, $data){
        if($this->last_opened_tag != NULL) #you don't need to store whitespace in between tags
            $this->data .= $data;
    }
    function close(&$parser, $tag){
        if($this->last_opened_tag == $tag){
            $this->parent = $this->data;
            $this->last_opened_tag = NULL;
        }
        array_pop($this->stack);
        if($this->stack) $this->parent = &$this->stack[count($this->stack)-1];
    }
}
function count_numeric_items(&$array){
    return is_array($array) ? count(array_filter(array_keys($array), 'is_numeric')) : 0;
}
?>