PHP DOMDocument parse HTML - php

Question

I have the following HTML markup
<div contenteditable="true" class="text"></div>
<div contenteditable="true" class="text"></div>
<div style="display: block;" class="ui-draggable">
<img class='avatar' src=""/>
<p style="">
<img class='pic' src=""/><br>
<span class='fulltext' style="display:none"></span>
</p>-<span class='create'></span>
<a class='permalink' href=""></a>
</div>
<div contenteditable="true" class="text"></div>
<div style="display: block;" class="ui-draggable">
<img class='avatar' src=""/>
<p style="">
<img class='pic' src=""/><br>
<span class='fulltext' style="display:none"></span>
</p><span class='create'></span><a class='permalink' href=""></a>
</div>
The parent div's can be more.In order to parse the information and to insert it in the DB I'm using the following code -
$dom = new DOMDocument();
$dom->loadHTML($xml);
$xpath = new DOMXPath($dom);
$div = $xpath->query('//div');
$i=0;
$q=1;
foreach($div as $book) {
$attr = $book->getAttribute('class');
//if div contenteditable
if($attr == 'text') {
echo '</br>'.$book->nodeValue."</br>";
}
else {
$new = new DOMDocument();
$newxpath = new DOMXPath($new);
$avatar = $xpath->query("(//img[#class='avatar']/#src)[$q]");
$picture = $xpath->query("(//p/img[#class='pic']/#src)[$q]");
$fulltext = $xpath->query("(//p/span[#class='fulltext'])[$q]");
$permalink = $xpath->query("(//a[#class='permalink'])[$q]");
echo $permalink->item(0)->nodeValue; //date
echo $permalink->item(0)->getAttribute('href');
echo $fulltext->item(0)->nodeValue;
echo $avatar->item(0)->value;
echo $picture->item(0)->value;
$q++;
}
$i++;
}
But I think that there's a better way for parsing the HTML. Is there? Thank you in advance

Answer 1

Note that DOMXPath::query supports a second param called contextparam. Also you won't need a second DOMDocument and DOMXPath inside the loop. Use:
$avatar = $xpath->query("img[#class='avatar']/#src", $book);
to get <img src=""> attribute nodes relative to the div nodes. If you follow my advices your example should be fine.
Here comes a version of your code that follows the above said:
$dom = new DOMDocument();
$dom->loadHTML($xml);
$xpath = new DOMXPath($dom);
$divs = $xpath->query('//div');
foreach($divs as $book) {
$attr = $book->getAttribute('class');
if($attr == 'text') {
echo '</br>'.$book->nodeValue."</br>";
} else {
$avatar = $xpath->query("img[#class='avatar']/#src", $book);
$picture = $xpath->query("p/img[#class='pic']/#src", $book);
$fulltext = $xpath->query("p/span[#class='fulltext']", $book);
$permalink = $xpath->query("a[#class='permalink']", $book);
echo $permalink->item(0)->nodeValue; //date
echo $permalink->item(0)->getAttribute('href');
echo $fulltext->item(0)->nodeValue;
echo $avatar->item(0)->value;
echo $picture->item(0)->value;
}
}

Answer 2

As a matter of fact, you do it the right way : html has to be parsed with a DOM object.
Then some optimisation can be brough :
$div = $xpath->query('//div');
is quite greedy, a getElementsByTagName should be more appropriate :
$div = $dom->getElementsByTagName('div');