Repeating XML parsing with PHP - php

Question

I asked before, and got the answers for one instance, now i have to parse multiple repeating instances of XML file into PHP variables; XML file looks like this:
<status>
<client type="s" name="root" desc="" protocol="server" protocolext="" au="0" thid="0x15e9190">
<request="0000" srvid="0000" time="" history="" answered=""></request>
<times login="2013-04-16T10:59:16+0200" online="7001" idle="0"></times>
<connection ip="127.0.0.1" port="0">OK</connection>
</client>
<client type="p" name="user1" desc="" protocol="run1" protocolext="" au="-1" thid="0x15f1790">
<request="0000" srvid="0000" time="" history="2667" answered=""></request>
<times login="2013-04-16T10:59:16+0200" online="7001" idle="6999"></times>
<connection ip="127.0.2.2" port="10002">CONNECTED</connection>
</client>
<client type="p" name="user2" desc="" protocol="run2" protocolext="" au="-1" thid="0x15f32b0">
<request="0000" srvid="0000" time="" history="" answered=""></request>
<times login="2013-04-16T10:59:16+0200" online="7001" idle="7001"></times>
<connection ip="127.0.3.1" port="12001">CONNECTED</connection>
</client>
<client type="p" name="user3" desc="" protocol="run1" protocolext="" au="-1" thid="0x1631170">
<request="0000" srvid="0000" time="" history="" answered=""></request>
<times login="2013-04-16T10:59:16+0200" online="7001" idle="7001"></times>
<connection ip="127.0.4.1" port="9600">CONNECTED</connection>
</client>
</status>
when i do it with Xpath, it works OK, but fetches only first data part into variables;
$dom = new DOMDocument();
$dom->loadXml($xml);
$xpath = new DOMXpath($dom);
$client_type = $xpath->evaluate('string(/status/client/#type)');
$name = $xpath->evaluate('string(/status/client/#name)');;
$conn_ip = $xpath->evaluate('string(/status/client/connection/#ip)');
and echoing variables:
echo $client_type;
echo $name ;
echo $conn_ip;
it returns only first values:
What would be the best way to pull ALL DATA from the file above?

Answer 1

Getting all the <client> nodes and then looping over them is a good way to make the dom traversal clearer. Here is an example of getting all the clients infos
$dom = new DOMDocument();
$dom->loadXml($xml);
$xpath = new DOMXpath($dom);
// use the double // to find ALL clients in the document
$clientXpath = "//client";
$clients = $xpath->evaluate($clientXpath);
// foreach client node
foreach ($clients as $ii=>$client) {
// get the type attribute of client node
echo $client->getAttribute('type') . "\n";
// get the name attribute of client node
echo $client->getAttribute('name') . "\n";
// get clients children
$children = $client->childNodes;
foreach ($children as $child) {
// ignore textnodes
if ($child instanceof DomText) {
continue;
}
// now concern ourself only with the connection tag, which
// contains the ip
if ($child->tagName == 'connection') {
print $child->getAttribute('ip') . "\n";
}
}
}

Answer 2

Xpath expression used in DOMXpath, return a DOMNodelist unless the result is casted to a scalar. So the example can be expanded to iterate over a list of nodes. Inside the loop the node is used as a context for the expressions.
$dom = new DOMDocument();
$dom->loadXml($xml);
$xpath = new DOMXpath($dom);
foreach ($xpath->evaluate('/status/client') as $client) {
var_dump(
$xpath->evaluate('string(#type)', $client),
$xpath->evaluate('string(#name)', $client),
$xpath->evaluate('string(connection/#ip)', $client)
);
}
Demo: https://eval.in/125083