PHP 8.3.0 RC 6 available for testing
    Submit a Pull Request Report a Bug

    xml_parse

    (PHP 4, PHP 5, PHP 7, PHP 8)

    xml_parse — 开始解析 XML 文档

    说明

    xml_parse(XMLParser $parser, string $data, bool $is_final = false): int

    xml_parse() 解析 XML 文档。已配置事件的处理程序根据需要被无限次调用。

    参数

    parser

    指向将要使用的 XML 解析器的指针。

    data

    要解析的数据块。可以通过使用新数据多次调用 xml_parse() 来分段解析文档，只要在解析最后一个数据时设置 is_final 参数且为 true 即可。

    is_final

    如果被设置为 true，则 data 为当前解析中最后一段数据。

    返回值

    成功时返回 1，失败时返回 0。

    若解析失败，可以使用如下函数获取错误信息：xml_get_error_code()、xml_error_string()、xml_get_current_line_number()、xml_get_current_column_number() 和 xml_get_current_byte_index()。

    注意:

    将 is_final 参数设置为 true，一些错误（例如 entity 错误）将会在数据末尾报告。

    更新日志

    版本 说明
    8.0.0 parser 现在接受 XMLParser 实例；之前接受有效的 xml resource。

    示例

    示例 #1 分块解析大型 XML 文档

    此示例展示了如何以块的形式读取和解析大型 XML 文档，这样就不必将整个文档保存在内存中。为简洁起见，省略了错误处理。

    <?php
    $stream     = fopen('large.xml', 'r');
    $parser = xml_parser_create();
    // set up the handlers here
    while (($data = fread($stream, 16384))) {
    xml_parse($parser, $data); // parse the current chunk
    }
    xml_parse($parser, '', true); // finalize parsing
    xml_parser_free($parser);
    fclose($stream);
    ＋add a note

    User Contributed Notes 19 notes

    up
    24
    neoyahuu at yahoo dot com
    15 years ago
    Instead of passing a URL, we can pass the XML content to this class (either you
    want to use CURL, Socks or fopen to retrieve it first) and instead of using
    array, I'm using separator '|' to identify which data to get (in order to make
    it short to retrieve a complex XML data). Here is my class with built-in fopen
    which you can pass URL or you can pass the content instead :

    p/s : thanks to this great help page.

    <?php

    class xx_xml {

    // XML parser variables
    var $parser;
    var     $name;
    var     $attr;
    var     $data = array();
    var     $stack = array();
    var     $keys;
    var     $path;

    // either you pass url atau contents.
    // Use 'url' or 'contents' for the parameter
    var $type;

    // function with the default parameter value
    function xx_xml($url='http://www.example.com', $type='url') {
    $this->type = $type;
    $this->url = $url;
    $this->parse();
    }

    // parse XML data
    function parse()
    {
    $data = '';
    $this->parser = xml_parser_create();
    xml_set_object($this->parser, $this);
    xml_set_element_handler($this->parser, 'startXML', 'endXML');
    xml_set_character_data_handler($this->parser, 'charXML');

    xml_parser_set_option($this->parser, XML_OPTION_CASE_FOLDING, false);

    if (    $this->type == 'url') {
    // if use type = 'url' now we open the XML with fopen

    if (!($fp = @fopen($this->url, 'rb'))) {
    $this->error("Cannot open {$this->url}");
    }

    while ((    $data = fread($fp, 8192))) {
    if (!    xml_parse($this->parser, $data, feof($fp))) {
    $this->error(sprintf('XML error at line %d column %d',
    xml_get_current_line_number($this->parser),
    xml_get_current_column_number($this->parser)));
    }
    }
    } else if (    $this->type == 'contents') {
    // Now we can pass the contents, maybe if you want
    // to use CURL, SOCK or other method.
    $lines = explode("\n",$this->url);
    foreach (    $lines as $val) {
    if (    trim($val) == '')
    continue;
    $data = $val . "\n";
    if (!    xml_parse($this->parser, $data)) {
    $this->error(sprintf('XML error at line %d column %d',
    xml_get_current_line_number($this->parser),
    xml_get_current_column_number($this->parser)));
    }
    }
    }
    }

    function     startXML($parser, $name, $attr) {
    $this->stack[$name] = array();
    $keys = '';
    $total = count($this->stack)-1;
    $i=0;
    foreach (    $this->stack as $key => $val) {
    if (    count($this->stack) > 1) {
    if (    $total == $i)
    $keys .= $key;
    else
    $keys .= $key . '|'; // The saparator
    }
    else
    $keys .= $key;
    $i++;
    }
    if (    array_key_exists($keys, $this->data)) {
    $this->data[$keys][] = $attr;
    } else
    $this->data[$keys] = $attr;
    $this->keys = $keys;
    }

    function     endXML($parser, $name) {
    end($this->stack);
    if (    key($this->stack) == $name)
    array_pop($this->stack);
    }

    function     charXML($parser, $data) {
    if (    trim($data) != '')
    $this->data[$this->keys]['data'][] = trim(str_replace("\n", '', $data));
    }

    function     error($msg) {
    echo     "<div align=\"center\">
    <font color=\"red\"><b>Error:     $msg</b></font>
    </div>"    ;
    exit();
    }
    }

    ?>

    And example of retrieving XML data:
    p/s: example use to retrieve weather

    <?php
    include_once "xx_xml.class.php";

    // Im using simple curl (the original is in class) to get the contents

    $pageurl = "http://xml.weather.yahoo.com/forecastrss?p=MYXX0008&u=c";
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt ($ch, CURLOPT_URL, $pageurl );
    $thecontents = curl_exec ( $ch );
    curl_close($ch);

    // We want to pass only a ready XML content instead of URL
    // But if you want to use URL , skip the curl functions above and use this
    // $xx4 = new xx_xml("url here",'url');

    $xx4 = new xx_xml($thecontents,'contents');
    // As you can see, we use saparator '|' instead of long array
    $Code = $xx4->data ['rss|channel|item|yweather:condition']['code'] ;
    $Celcius = $xx4->data ['rss|channel|item|yweather:condition']['temp'] ;
    $Text = $xx4->data ['rss|channel|item|yweather:condition']['text'] ;
    $Cityname = $xx4->data ['rss|channel|yweather:location']['city'] ;

    ?>

    Hope this helps.
    up
    7
    lz_speedy at web dot de
    14 years ago
    Best seen xml2array function ever
    <?php
    function xml2array($url, $get_attributes = 1, $priority = 'tag')
    {
    $contents = "";
    if (!    function_exists('xml_parser_create'))
    {
    return array ();
    }
    $parser = xml_parser_create('');
    if (!(    $fp = @ fopen($url, 'rb')))
    {
    return array ();
    }
    while (!    feof($fp))
    {
    $contents .= fread($fp, 8192);
    }
    fclose($fp);
    xml_parser_set_option($parser, XML_OPTION_TARGET_ENCODING, "UTF-8");
    xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, 0);
    xml_parser_set_option($parser, XML_OPTION_SKIP_WHITE, 1);
    xml_parse_into_struct($parser, trim($contents), $xml_values);
    xml_parser_free($parser);
    if (!    $xml_values)
    return;     //Hmm...
    $xml_array = array ();
    $parents = array ();
    $opened_tags = array ();
    $arr = array ();
    $current = & $xml_array;
    $repeated_tag_index = array ();
    foreach (    $xml_values as $data)
    {
    unset (    $attributes, $value);
    extract($data);
    $result = array ();
    $attributes_data = array ();
    if (isset (    $value))
    {
    if (    $priority == 'tag')
    $result = $value;
    else
    $result['value'] = $value;
    }
    if (isset (    $attributes) and $get_attributes)
    {
    foreach (    $attributes as $attr => $val)
    {
    if (    $priority == 'tag')
    $attributes_data[$attr] = $val;
    else
    $result['attr'][$attr] = $val; //Set all the attributes in a array called 'attr'
    }
    }
    if (    $type == "open")
    {
    $parent[$level -1] = & $current;
    if (!    is_array($current) or (!in_array($tag, array_keys($current))))
    {
    $current[$tag] = $result;
    if (    $attributes_data)
    $current[$tag . '_attr'] = $attributes_data;
    $repeated_tag_index[$tag . '_' . $level] = 1;
    $current = & $current[$tag];
    }
    else
    {
    if (isset (    $current[$tag][0]))
    {
    $current[$tag][$repeated_tag_index[$tag . '_' . $level]] = $result;
    $repeated_tag_index[$tag . '_' . $level]++;
    }
    else
    {
    $current[$tag] = array (
    $current[$tag],
    $result
    );
    $repeated_tag_index[$tag . '_' . $level] = 2;
    if (isset (    $current[$tag . '_attr']))
    {
    $current[$tag]['0_attr'] = $current[$tag . '_attr'];
    unset (    $current[$tag . '_attr']);
    }
    }
    $last_item_index = $repeated_tag_index[$tag . '_' . $level] - 1;
    $current = & $current[$tag][$last_item_index];
    }
    }
    elseif (    $type == "complete")
    {
    if (!isset (    $current[$tag]))
    {
    $current[$tag] = $result;
    $repeated_tag_index[$tag . '_' . $level] = 1;
    if (    $priority == 'tag' and $attributes_data)
    $current[$tag . '_attr'] = $attributes_data;
    }
    else
    {
    if (isset (    $current[$tag][0]) and is_array($current[$tag]))
    {
    $current[$tag][$repeated_tag_index[$tag . '_' . $level]] = $result;
    if (    $priority == 'tag' and $get_attributes and $attributes_data)
    {
    $current[$tag][$repeated_tag_index[$tag . '_' . $level] . '_attr'] = $attributes_data;
    }
    $repeated_tag_index[$tag . '_' . $level]++;
    }
    else
    {
    $current[$tag] = array (
    $current[$tag],
    $result
    );
    $repeated_tag_index[$tag . '_' . $level] = 1;
    if (    $priority == 'tag' and $get_attributes)
    {
    if (isset (    $current[$tag . '_attr']))
    {
    $current[$tag]['0_attr'] = $current[$tag . '_attr'];
    unset (    $current[$tag . '_attr']);
    }
    if (    $attributes_data)
    {
    $current[$tag][$repeated_tag_index[$tag . '_' . $level] . '_attr'] = $attributes_data;
    }
    }
    $repeated_tag_index[$tag . '_' . $level]++; //0 and 1 index is already taken
    }
    }
    }
    elseif (    $type == 'close')
    {
    $current = & $parent[$level -1];
    }
    }
    return (    $xml_array);
    }
    ?>

    Returns a well formed array like the structure of the xml-document

    <root>
    <child1>
    <child1child1/>
    </child1>
    </root>

    create an array like
    array[root][child1][child1child1]

    lg
    up
    2
    Kyle Bresin
    17 years ago
    Just wanted to note a small bug in bbellwfu's class (which is really great btw).

    It fails to capture any datums which are equal to numerical zero.

    The problem lies in the function tagData, the first if statement should be:

    if(trim($tagData) != '') {
    up
    2
    Ashok dot 893 at gmail dot com
    13 years ago
    This is very simple way to convert all applicable objects into associative array. This works with not only SimpleXML but any kind of object. The input can be either array or object. This function also takes an options parameter as array of indices to be excluded in the return array. And keep in mind, this returns only the array of non-static and accessible variables of the object since using the function get_object_vars().

    <?php
    function objectsIntoArray($arrObjData, $arrSkipIndices = array())
    {
    $arrData = array();

    // if input is object, convert into array
    if (is_object($arrObjData)) {
    $arrObjData = get_object_vars($arrObjData);
    }

    if (    is_array($arrObjData)) {
    foreach (    $arrObjData as $index => $value) {
    if (    is_object($value) || is_array($value)) {
    $value = objectsIntoArray($value, $arrSkipIndices); // recursive call
    }
    if (    in_array($index, $arrSkipIndices)) {
    continue;
    }
    $arrData[$index] = $value;
    }
    }
    return     $arrData;
    }
    ?>

    Usage:

    <?php
    $xmlUrl     = "feed.xml"; // XML feed file/URL
    $xmlStr = file_get_contents($xmlUrl);
    $xmlObj = simplexml_load_string($xmlStr);
    $arrXml = objectsIntoArray($xmlObj);
    print_r($arrXml);
    ?>
    up
    1
    php at b635 dot com
    16 years ago
    The suggestions below have been a great help, but there was one thing I really needed...

    I'm parsing Amazon XML data, and I wanted to be able to index into the array using something like:

    <?php
    print "<p>" . $strAXML->arrOutput[ITEMLOOKUPRESPONSE][ITEMS][ITEM][SMALLIMAGE][URL];
    ?>

    To solve this, I had to push all of the open tags onto a separate stack, add any tag data to the tail end of an attributed multi-dimensional array, and then pop the tag name off of the stack once it was closed...

    <?php

    class xml2array {

    var     $arrOutput = array();
    var     $arrName = array();
    var     $objParser;
    var     $strXmlData;

    function     parse($strInputXML) {

    // standard XML parse object setup

    $this->objParser = xml_parser_create ();
    xml_set_object($this->objParser,$this);
    xml_set_element_handler($this->objParser, "tagOpen", "tagClosed");

    xml_set_character_data_handler($this->objParser, "tagData");

    $this->strXmlData = xml_parse($this->objParser,$strInputXML );
    if(!    $this->strXmlData) {
    die(    sprintf("XML error: %s at line %d",
    xml_error_string(xml_get_error_code($this->objParser)),
    xml_get_current_line_number($this->objParser)));
    }

    xml_parser_free($this->objParser);

    return     $this->arrOutput;
    }

    function     tagOpen($parser, $name, $attrs) {
    // push the current tag name to an array of still-open tag names
    array_push ($this->arrName, $name);

    // merge the array of current attributes to the open tag
    // NOTE: this does not currently handle multiple attributes with the same name
    // (i.e. it will overwrite them with the last values)

    $strEval = "\$this->arrOutput";
    foreach (    $this->arrName as $value) {
    $strEval .= "[" . $value . "]";
    }
    $strEval = $strEval . " = array_merge (" . $strEval . ",\$attrs);";

    eval (    $strEval);
    }

    function     tagData($parser, $tagData) {

    // set the latest open tag equal to the tag data

    $strEval = "\$this->arrOutput";
    foreach (    $this->arrName as $value) {
    $strEval .= "[" . $value . "]";
    }

    $strEval = $strEval . " = \$tagData;";

    eval (    $strEval);
    }


    function     tagClosed($parser, $name) {

    // pop this tag (and any subsequent tags) off the stack of open tag names

    for ($i = count ($this->arrName) - 1; $i > 0; $i--) {
    $currName = $this->arrName[$i];
    array_pop ($this->arrName);
    if (    $currName == $name) {
    break;
    }
    }
    }

    }

    ?>
    up
    1
    bbellwfu at gmail dot com
    18 years ago
    Just improving a little bit on the code examples from tgrabietz and randlem below... everything in one pretty class, plus some checks in place so that the element data doesnt get split up (thanks to flobee on the xml_set_character_data_handler page)

    <?php

    /* Usage
    Grab some XML data, either from a file, URL, etc. however you want. Assume storage in $strYourXML;

    $objXML = new xml2Array();
    $arrOutput = $objXML->parse($strYourXML);
    print_r($arrOutput); //print it out, or do whatever!

    */
    class xml2Array {

    var     $arrOutput = array();
    var     $resParser;
    var     $strXmlData;

    function     parse($strInputXML) {

    $this->resParser = xml_parser_create ();
    xml_set_object($this->resParser,$this);
    xml_set_element_handler($this->resParser, "tagOpen", "tagClosed");

    xml_set_character_data_handler($this->resParser, "tagData");

    $this->strXmlData = xml_parse($this->resParser,$strInputXML );
    if(!    $this->strXmlData) {
    die(    sprintf("XML error: %s at line %d",
    xml_error_string(xml_get_error_code($this->resParser)),
    xml_get_current_line_number($this->resParser)));
    }

    xml_parser_free($this->resParser);

    return     $this->arrOutput;
    }
    function     tagOpen($parser, $name, $attrs) {
    $tag=array("name"=>$name,"attrs"=>$attrs);
    array_push($this->arrOutput,$tag);
    }

    function     tagData($parser, $tagData) {
    if(    trim($tagData)) {
    if(isset(    $this->arrOutput[count($this->arrOutput)-1]['tagData'])) {
    $this->arrOutput[count($this->arrOutput)-1]['tagData'] .= $tagData;
    }
    else {
    $this->arrOutput[count($this->arrOutput)-1]['tagData'] = $tagData;
    }
    }
    }

    function     tagClosed($parser, $name) {
    $this->arrOutput[count($this->arrOutput)-2]['children'][] = $this->arrOutput[count($this->arrOutput)-1];
    array_pop($this->arrOutput);
    }
    }
    ?>

    Will output something like...

    <snippet>
    Array
    (
    [0] => Array
    (
    [name] => GETMESSAGESRESPONSE
    [attrs] => Array
    (
    )

    [children] => Array
    (
    [0] => Array
    (
    [name] => STATUS
    [attrs] => Array
    (
    )

    )

    </snippet>
    up
    2
    alex dot garcia at noos dot fr
    18 years ago
    Here is the inverse function which takes parsed xml array in entry and outputs xml string

    enjoy !

    function getXmlFromArray($root){

    if(count($root) > 0){

    $curr_name = $root['name'];
    $attribs = $root['attrs'];
    $curr_childs = $root['children'];
    $curr_data = $root['cdata'];

    $xml .= '<'.$curr_name;

    if(count($attribs) > 0){
    $i = 1;
    foreach($attribs as $key => $value){
    $curr_attribs .= $key.'="'.$value.'"';
    $i++;
    if($i <= count($attribs)){
    $curr_attribs .= ' ';
    }
    }
    $xml .= ' '.$curr_attribs;
    }

    if($curr_data != ''){
    $xml .= '><![CDATA['.$curr_data.']]></'.$curr_name.'>';
    } else {
    if(count($curr_childs) > 0){
    $xml .= '>';
    foreach($curr_childs as $child){
    $xml .= getXmlFromArray($child);
    }
    $xml .= '</'.$curr_name.'>';
    } else {
    $xml .= '/>';
    }
    }

    }
    return $xml;
    }
    up
    0
    ben at autonomic dot net
    17 years ago
    bbellwfu's code does not handle 'text nodes' properly.
    Consider the innards of a tag like <root>xxx<tag2/>yyy</root>
    The 'tagData' for root will be "xxxyyy" and you have lost all information about where "tag2" was in that sequence.

    Quick and dirty hack.

    Replace tagData with this code :
    function tagData($parser, $tagData) {
    $last_element=count($this->arrOutput)-1;
    $this->arrOutput[$last_element]['children'][] = array("textnode",$tagData);
    }

    What this does is adds 'textnodes' as children of its containing parent, *in the right sequence* (rather like the internet browsers do it). This then lets you do some more sensible secondary work like recursively looking up internal references within the document...
    up
    0
    tim at alloutinteraction dot com
    18 years ago
    I wanted to create a really simple XML parser, but I found the array management in xml_parse a bit daunting. So I flattened my XML and parsed it using string matching. It wouldn't be difficult to add xml depth (of 2 plus levels) by modifying the parsedXML array.

    <?
    // here's the raw html
    $xmlRaw="<order>Order data</order><label>Label data</label><control>123</control>";
    // here are the xml field names
    $xmlFieldNames=array("order", "label", "control");
    // for each xml field...
    foreach ($xmlFieldNames as $xmlField) {
    if(strpos($xmlRaw,$xmlField)!==false){
    // I've broken 1 single line into 4 for display purposes
    $parsedXML[$xmlField]=substr($xmlRaw,
    strpos($xmlRaw,"<$xmlField>")+strlen("<$xmlField>"),
    strpos($xmlRaw,"</$xmlField>")-strlen("<$xmlField>")
    -strpos($xmlRaw,"<$xmlField>"));
    }
    }
    print_r($parsedXML);
    // prints: Array ( [order] => Order data [label] => Label data [control] => 123 )

    ?>

    Hope you find this useful (coded it while ill in bed with streaming cold, but felt much better afterwards!)

    Tim (a lazy coder)
    up
    0
    ByK
    19 years ago
    modified from yours code. I think it's work!!.
    class CXml
    {
    var $xml_data;
    var $obj_data;
    var $pointer;

    function CXml() { }

    function Set_xml_data( &$xml_data )
    {
    $this->index = 0;
    $this->pointer[] = &$this->obj_data;

    //strip white space between tags
    $this->xml_data = eregi_replace(">"."[[:space:]]+"."<","><",$xml_data);
    $this->xml_parser = xml_parser_create( "UTF-8" );

    xml_parser_set_option( $this->xml_parser, XML_OPTION_CASE_FOLDING, false );
    xml_set_object( $this->xml_parser, &$this );
    xml_set_element_handler( $this->xml_parser, "_startElement", "_endElement");
    xml_set_character_data_handler( $this->xml_parser, "_cData" );

    xml_parse( $this->xml_parser, $this->xml_data, true );
    xml_parser_free( $this->xml_parser );
    }

    function _startElement( $parser, $tag, $attributeList )
    {
    foreach( $attributeList as $name => $value )
    {
    $value = $this->_cleanString( $value );
    $object->$name = $value;
    }
    //replaces the special characters with the underscore (_) in tag name
    $tag = preg_replace("/[:\-\. ]/", "_", $tag);
    eval( "\$this->pointer[\$this->index]->" . $tag . "[] = \$object;" );
    eval( "\$size = sizeof( \$this->pointer[\$this->index]->" . $tag . " );" );
    eval( "\$this->pointer[] = &\$this->pointer[\$this->index]->" . $tag . "[\$size-1];" );

    $this->index++;
    }

    function _endElement( $parser, $tag )
    {
    array_pop( $this->pointer );
    $this->index--;
    }

    function _cData( $parser, $data )
    {
    if (empty($this->pointer[$this->index])) {
    if (rtrim($data, "\n"))
    $this->pointer[$this->index] = $data;
    } else {
    $this->pointer[$this->index] .= $data;
    }
    }

    function _cleanString( $string )
    {
    return utf8_decode( trim( $string ) );
    }
    }

    $m_xml = new CXml();
    $xml_data = file_get_contents( $filename );

    $m_xml->Set_XML_data( $xml_data );

    $newsid = $m_xml->obj_data->root[0]->NewsID[0];
    up
    0
    michelek
    20 years ago
    its maybe not better, but me thinks its more stright-forward

    --INPUT:

    <?xml version="1.0" encoding="UTF-8"?>
    <world>
    <country name="sweden">
    <city name="stockholm">
    <user>Adam</user>
    <user>Eva</user>
    </city>
    <city name="g?teborg">
    <user>God</user>
    </city>
    </country>
    <country name="usa">
    <city name="new york">
    <user>Clinton</user>
    <user>Bush</user>
    </city>
    </country>
    </world>

    --CODE:
    <?
    /**
    * m_i_h_k_e_l_AT_w_w_DOT_e_e
    * 26.10.2003
    **/
    $filename = "m.m.xml";
    $xmlC = new XmlC();
    $xml_data = file_get_contents( $filename );

    $xmlC->Set_XML_data( $xml_data );

    echo( "<pre>\n" );
    print_r( $xmlC->obj_data );
    echo( "</pre>\n" );

    class XmlC
    {
    var $xml_data;
    var $obj_data;
    var $pointer;

    function XmlC()
    {
    }

    function Set_xml_data( &$xml_data )
    {
    $this->index = 0;
    $this->pointer[] = &$this->obj_data;

    $this->xml_data = $xml_data;
    $this->xml_parser = xml_parser_create( "UTF-8" );

    xml_parser_set_option( $this->xml_parser, XML_OPTION_CASE_FOLDING, false );
    xml_set_object( $this->xml_parser, &$this );
    xml_set_element_handler( $this->xml_parser, "_startElement", "_endElement");
    xml_set_character_data_handler( $this->xml_parser, "_cData" );

    xml_parse( $this->xml_parser, $this->xml_data, true );
    xml_parser_free( $this->xml_parser );
    }

    function _startElement( $parser, $tag, $attributeList )
    {
    foreach( $attributeList as $name => $value )
    {
    $value = $this->_cleanString( $value );
    $object->$name = $value;
    }
    eval( "\$this->pointer[\$this->index]->" . $tag . "[] = \$object;" );
    eval( "\$size = sizeof( \$this->pointer[\$this->index]->" . $tag . " );" );
    eval( "\$this->pointer[] = &\$this->pointer[\$this->index]->" . $tag . "[\$size-1];" );

    $this->index++;
    }

    function _endElement( $parser, $tag )
    {
    array_pop( $this->pointer );
    $this->index--;
    }

    function _cData( $parser, $data )
    {
    if( trim( $data ) )
    {
    $this->pointer[$this->index] = trim( $data );
    }
    }

    function _cleanString( $string )
    {
    return utf8_decode( trim( $string ) );
    }

    }
    ?>
    up
    -1
    Programmerdude
    9 years ago
    Due to restrictions in libxml2 the maximum length of $data is 9.5MB. If you need to process more than 9.5MB you can do so by calling the function multiple times and setting $is_final on the last call.
    up
    -1
    dgrimes at scvl dot com
    16 years ago
    One note about magic quotes: magic_quotes_runtime needs to be disabled before parsing XML. It can cause strange errors during parsing. Just add the following at the top of your program:

    set_magic_quotes_runtime(0);

    If you need magic quotes you can use stripslashes or save the current magic quotes setting with get_magic_quotes_runtime() then disable and parse your XML and then restore the previous magic quotes setting.
    up
    -1
    tgrabietz at bupnet dot de
    19 years ago
    it's like randlem at gmail dot com's great code, without using a "class container" but parsing cdata. The script returns the tree-structure in a single array.

    <?php
    $file     = 'simple.xml';
    $stack = array();

    function     startTag($parser, $name, $attrs)
    {
    global     $stack;
    $tag=array("name"=>$name,"attrs"=>$attrs);
    array_push($stack,$tag);

    }

    function     cdata($parser, $cdata)
    {
    global     $stack,$i;

    if(    trim($cdata))
    {
    $stack[count($stack)-1]['cdata']=$cdata;
    }
    }

    function     endTag($parser, $name)
    {
    global     $stack;
    $stack[count($stack)-2]['children'][] = $stack[count($stack)-1];
    array_pop($stack);
    }

    $xml_parser = xml_parser_create();
    xml_set_element_handler($xml_parser, "startTag", "endTag");
    xml_set_character_data_handler($xml_parser, "cdata");

    $data = xml_parse($xml_parser,file_get_contents($file));
    if(!    $data) {
    die(    sprintf("XML error: %s at line %d",
    xml_error_string(xml_get_error_code($xml_parser)),
    xml_get_current_line_number($xml_parser)));
    }

    xml_parser_free($xml_parser);

    print(    "<pre>\n");
    print_r($stack);
    print(    "</pre>\n");
    ?>
    up
    -1
    alejandro dot anv at gmail dot com
    9 years ago
    Here is my array2xml function. I made it because I coudn't find a generic recursive function that generates a prety formatted XML from any array.

    Usage:

    <?php echo xml2array('rootname',$myarray); ?>

    or if you want to specify a spacing character(s) other than TAB:

    <?php echo xml2array('rootname',$myarray,' '); ?>

    or if you want avoid spacing at all

    <?php echo xml2array('rootname',$myarray,''); ?>

    <?php
    function array2xml($nombre,$valor,$tab="\t",$nivel=0){

    if (    $nivel==0) $r='<?xml version="1.0" encoding="ISO-8859-1"?>';

    $arr=!is_numeric($nombre);
    $spacing=str_repeat($tab,$nivel);
    if (    $tab!='') $newline="\n";
    else     $newline='';

    if (    is_array($valor)){

    if (    count($valor)>0){
    $r.=$newline;
    //$r.=str_repeat($tab,$nivel);
    foreach($valor as $k=>$v){
    if (    $arr) $r.="$spacing<$nombre>";

    if (    is_array($v)) $r.=array2xml($k,$v,$tab,$nivel+$arr);
    else     $r.="$spacing<$k>$v</$k>$newline";

    if (    $arr) $r.="$spacing</$nombre>$newline";
    }    //foreach
    }//count
    else $r.="$spacing<$nombre></$nombre>"; // it's an empty array
    }
    else {
    if (    $arr) $r.="$spacing<$nombre>";
    $r.=$valor;
    if (    $arr) $r.="</$nombre>$newline";
    }

    return     $r;
    }
    ?>
    up
    -3
    Adam Tylmad
    20 years ago
    I've created a parser that returns an
    object based on a xml document.

    example:
    <?xml version="1.0" encoding="ISO-8859-1" ?>
    <country name="sweden">
    <city name="stockholm">
    <user>Adam</user>
    <user>Eve</user>
    </city>
    <city name="g?teborg">
    <user>God</user>
    </city>
    </country>
    <country name="usa">
    <city name="new york">
    <user>Clinton</user>
    <user>Bush</user>
    </city>
    </country>

    generates the following object structure:
    [country] => Array
    (
    [0] => stdClass Object
    (
    [name] => sweden
    [city] => Array
    (
    [0] => stdClass Object
    (
    [name] => stockholm
    [user] => Array
    (
    [0] => Adam
    [1] => Eve
    )
    )
    [1] => stdClass Object
    (
    [name] => g?teborg
    [user] => God
    )
    )
    )
    [1] => stdClass Object
    (
    [name] => usa
    [city] => stdClass
    (
    [name] => new york
    [user] => Array
    (
    [0] => Clinton
    [1] => Bush
    )
    )
    )
    )

    Here is the code:

    class XMLParser {
    var $path;
    var $result;

    function XMLParser($encoding, $data) {
    $this->path = "\$this->result";
    $this->index = 0;

    $xml_parser = xml_parser_create($encoding);
    xml_set_object($xml_parser, &$this);
    xml_set_element_handler($xml_parser, 'startElement', 'endElement');
    xml_set_character_data_handler($xml_parser, 'characterData');

    xml_parse($xml_parser, $data, true);
    xml_parser_free($xml_parser);
    }

    function startElement($parser, $tag, $attributeList) {
    eval("\$vars = get_object_vars(".$this->path.");");
    $this->path .= "->".$tag;
    if ($vars and array_key_exists($tag, $vars)) {
    eval("\$data = ".$this->path.";");
    if (is_array($data)) {
    $index = sizeof($data);
    $this->path .= "[".$index."]";
    } else if (is_object($data)) {
    eval($this->path." = array(".$this->path.");");
    $this->path .= "[1]";
    }
    }
    eval($this->path." = null;");

    foreach($attributeList as $name => $value)
    eval($this->path."->".$name. " = '".XMLParser::cleanString($value)."';");
    }

    function endElement($parser, $tag) {
    $this->path = substr($this->path, 0, strrpos($this->path, "->"));
    }

    function characterData($parser, $data) {
    eval($this->path." = '".trim($data)."';");
    }
    }

    enjoy! And please make it better if you can ;-)
    up
    -1
    joris dot landman at chello dot nl
    16 years ago
    This page has been a great help! I've adapted the examples below to make a class to parse an X(HT)ML file to a multidimensional array.

    Spaces, tabs, breaks, etc. are included in the array as TEXT_NODE, since in XHTML they may be functional. A function to trim them can easily be added if so desired.

    I've written my class to store multiple tags with the same name. In order to do this I've used nested arrays with the key NODES to store the order in which tags and data were parsed. These NODES arrays can be used as a blueprint to reconstruct the X(HT)ML part of the document in it's entirity, including formatting. (Doctype will have to be added for validity).

    <?php

    class my_xml_object {

    var     $xml_data; # raw xml data from file
    function parse_xml_file($my_uri) {
    $this->xml_data = null; # clear previously parsed file and related variables
    if (is_file($my_uri) && is_readable($my_uri)) { # existing and readable uri?
    $my_file = fopen($my_uri, "r");
    while(    $my_xml_input = fread($my_file, filesize($my_uri))) {
    $this->xml_data .= $my_xml_input; # add data to xml_data
    }
    fclose($my_file);
    $this->parse_xml_data($this->xml_data); # parse data
    } else {
    trigger_error("supplied argument is not a URI to a (readable) file", E_USER_ERROR);
    }
    }

    var     $xml_array = Array(); # xml array from parsed data
    function parse_xml_data($my_data) { # adapted from class by randlem@gmail.com, tgrabietz@bupnet.de, bbellwfu@gmail.com, Kyle Bresin - see http://nl2.php.net/xml_parse
    $this->xml_array = Array(0 => Array()); # clear previously parsed file and related variables; populate first element
    $my_parser = xml_parser_create(); # set up parser
    xml_set_object($my_parser, $this); # enable parser within object
    xml_set_element_handler($my_parser, "xml_tag_open", "xml_tag_close");
    xml_set_character_data_handler($my_parser, "xml_tag_data");
    if (!    xml_parse($my_parser, $my_data)) {
    trigger_error("data can not be parsed", E_USER_ERROR); # inspect problems #die(sprintf("<br />\n<b>Error</b>: %s on line <b>%d</b><br />\n", xml_error_string(xml_get_error_code($my_parser)), xml_get_current_line_number($my_parser)));
    }
    xml_parser_free($my_parser); # free parser
    return $this->xml_array; # return xml array
    }

    var     $my_branch = Array();
    function     xml_tag_open($my_parser, $my_name, $my_attributes) {
    array_push($this->my_branch, $my_name); # add tag name to branch
    $this->xml_array[] = Array(); # nest array in xml array for data
    if (count($my_attributes)) {
    $this->xml_array[count($this->xml_array) - 1]["ATTRIBUTES"] = $my_attributes; # nest attributes array
    }
    }

    function     xml_tag_data($my_parser, $my_data) {
    $this->xml_array[count($this->my_branch)]["TEXT_NODE"][] = $my_data; # add data to nested array
    $this->xml_array[count($this->my_branch)]["NODES"][] = "TEXT_NODE"; # add text node to nested NODES array
    }

    function     xml_tag_close($my_parser, $my_name) {
    $this->xml_array[count($this->my_branch) - 1][$this->my_branch[count($this->my_branch) - 1]][] = $this->xml_array[count($this->xml_array) - 1]; # nest arrays to follow document structure
    array_pop($this->xml_array); # pop off element that was nested
    $this->xml_array[count($this->my_branch) - 1]["NODES"][] = $my_name; # add tag node to nested NODES array
    array_pop($this->my_branch); # update branch
    }

    ?>
    up
    -1
    james @at@ mercstudio dot Com dot nospam
    17 years ago
    hi,

    i've modified bbellwfu at gmail dot com to as below:

    features added:
    - toXML (convert back array to xml string)
    - changed name, according to macromedia flash xml concept : children -> childrens, tagdata -> nodevalue, name -> nodename,
    - added pointer firstchild to childrens[0] (if exists)

    some findings that i would like to share:
    - <![cdata[my value here]]> (does not work on property value
    - xml file must be htmlentity based (if not using cdata)
    - xml line feed on node data seems to be double line feed on windows (still figuring why)
    - xml line feed on attribute value seems to be ignored...

    here's my code below :)

    class u007xml
    {
    var $arrOutput = array();
    var $resParser;
    var $strXmlData;


    function u007xml($tfile = "")
    {
    if(trim($tfile) != "") { $this->loadFile($tfile);}
    }

    function loadFile($tfile)
    {
    $this->thefile = $tfile;

    $th = file($tfile);
    $tdata = implode("\n", $th);

    return $this->parse($tdata);
    }

    function parse($strInputXML)
    {
    $this->resParser = xml_parser_create ();
    xml_set_object($this->resParser,$this);
    xml_set_element_handler($this->resParser, "tagOpen", "tagClosed");

    xml_set_character_data_handler($this->resParser, "tagData");

    $this->strXmlData = xml_parse($this->resParser,$strInputXML );

    if(!$this->strXmlData) {
    die(sprintf("XML error: %s at line %d",
    xml_error_string(xml_get_error_code($this->resParser)),
    xml_get_current_line_number($this->resParser)));
    }

    xml_parser_free($this->resParser);

    return $this->arrOutput;
    }

    //called on each xml tree
    function tagOpen($parser, $name, $attrs) {
    $tag=array("nodename"=>$name,"attributes"=>$attrs);
    array_push($this->arrOutput,$tag);
    }

    //called on data for xml
    function tagData($parser, $tagData) {
    if(trim($tagData)) {
    if(isset($this->arrOutput[count($this->arrOutput)-1]['nodevalue'])) {
    $this->arrOutput[count($this->arrOutput)-1]['nodevalue'] .= $this->parseXMLValue($tagData);
    }
    else {
    $this->arrOutput[count($this->arrOutput)-1]['nodevalue'] = $this->parseXMLValue($tagData);
    }
    }
    }

    //called when finished parsing
    function tagClosed($parser, $name) {
    $this->arrOutput[count($this->arrOutput)-2]['childrens'][] = $this->arrOutput[count($this->arrOutput)-1];

    if(count ($this->arrOutput[count($this->arrOutput)-2]['childrens'] ) == 1)
    {
    $this->arrOutput[count($this->arrOutput)-2]['firstchild'] =& $this->arrOutput[count($this->arrOutput)-2]['childrens'][0];
    }
    array_pop($this->arrOutput);
    }

    function toArray()
    {
    //not used, we can call loadString or loadFile instead...
    }


    function parseXMLValue($tvalue)
    {
    $tvalue = htmlentities($tvalue);
    return $tvalue;
    }

    function toXML($tob = null)
    {
    //return back xml
    $result = "";

    if( $tob == null)
    {
    $tob = $this->arrOutput;
    }

    if(!isset($tob))
    {
    echo "XML Array empty...";
    return null;
    }


    for($c = 0; $c < count($tob); $c++)
    {
    $result .="<" . $tob[$c]["nodename"];

    while (list($key, $value) = each($tob[$c]["attributes"]))
    {
    $result .=" " . $key."=\"" . $this->parseXMLValue($value) . "\"";
    }

    $result .= ">";

    //assign node value
    if( isset($tob[$c]["nodevalue"]) )
    {
    $result .= $tob[$c]["nodevalue"];
    }

    if( count($tob[$c]["childrens"]) > 0 )
    {
    $result .= "\r\n" . $this->toXML(&$tob[$c]["childrens"]) . "";
    }

    $result .= "</" . $tob[$c]["nodename"] . ">\r\n";


    }//end of each array...

    return $result;
    }

    function displayXML()
    {
    print_r($this->arrOutput);
    }

    function getXML($tob = null)
    {
    return "<?xml version='1.0'?>\r\n" . $this->toXML($tob);
    }

    }//end of u007xml class

    //examples below:

    $xx = new u007xml();
    $xx->loadFile("xml3.xml");

    //$xx->displayXML();

    print $xx->getXML();
    up
    -1
    randlem at gmail dot com
    19 years ago
    Here's a handy way to generate a tree that can be can be decended easily.

    <?php
    $file     = 'xmltest.xml';

    $tag_tree = array();
    $stack = array();

    class     tag {
    var     $name;
    var     $attrs;
    var     $children;
    function     tag($name, $attrs, $children) {
    $this->name = $name;
    $this->attrs = $attrs;
    $this->children = $children;
    }
    }

    function     startTag($parser, $name, $attrs) {
    global     $tag_tree, $stack;
    $tag = new tag($name,$attrs,'');
    array_push($stack,$tag);
    }

    function     endTag($parser, $name) {
    global     $stack;
    $stack[count($stack)-2]->children[] = $stack[count($stack)-1];
    array_pop($stack);
    }

    $xml_parser = xml_parser_create();
    xml_set_element_handler($xml_parser, "startTag", "endTag");

    $data = xml_parse($xml_parser,file_get_contents($file));
    if(!    $data) {
    die(    sprintf("XML error: %s at line %d",
    xml_error_string(xml_get_error_code($xml_parser)),
    xml_get_current_line_number($xml_parser)));
    }

    xml_parser_free($xml_parser);

    print(    "\n");
    print_r($stack);
    print(    "\n");
    ?>
    ＋add a note
    To Top