CODEWITHSUNDEEP
phpdomdocument
sharf
sharf

Reputation: 2143

Web scraper with DOMDocument

I'm trying to scrape a web page for content, using file_get_contents to grab the HTML and then using a DOMDocument object. My problem is that I cannot get the appropriate information. I'm not sure if this is because I'm using DOMDocument's methods wrong, or if the (X)HTML in my source is just poor.

In the source, there is an element with an id of 'cards', which has two child divs. I want the first child, which has many child divs, who in turn have an anchor child with div child. I want the href from the anchor and the nodeValue from it's child div.

The structure is like this:

<div id="cards">
    <div class="grid">
        <div class="card-wrap">
            <a href="linkValue">
                <img src="..."/>
                <div>nameValue</div>
            </a>
        </div>
        ...
   </div>
   <div id="...">
   </div>
</div>

I've started out with $cards = $dom->getElementById("cards"). I get a DOMText Object, a DOMElement Object, a DOMText Object, a DOMElement Object, and a DOMText Object. I then use $grid = $cards->childNodes->item(1) to get the first DOMElement Object, which is presumably the .grid element. However, when I then iterate through the $grid with:

foreach($grid->childNodes as $item){
    if($item->nodeName == "div"){
        echo $item->nodeName,' | ',$item->nodeValue,'<br>';
    }
}

I end up with a page full of "div | nameValue" where nameValue is the embedded div's nodeValue, and I am unable to locate the anchors to get their href value.

Am I doing something obviously wrong with my DOMDocument, or perhaps there is something more going on here?

Upvotes: 3

Views: 1142

Answers (2)

mhall
mhall

Reputation: 3701

The XPath way:

$src = <<<EOS
<div id="cards">
    <div class="grid">
        <div class="card-wrap">
            <a href="linkValue">
                <img src="..."/>
                <div>nameValue</div>
            </a>
        </div>
   </div>
   <div id="whatever">
   </div>
</div>
EOS;

$xml = new SimpleXMLElement($src);

list ($anchor) = $xml->xpath('//div[@id="cards"]/div[1]/div[1]/a');
echo $anchor->div, ' => ', $anchor['href'], PHP_EOL;

"Get anchor of first child div of first child div of div with an id of 'cards'"

Output:

nameValue => linkValue

Upvotes: 0

Don Scott
Don Scott

Reputation: 3457

Well, from your example code if($item->nodeName == "div"){ is very going to preclude any <a> tag. Additionally, I do not believe childNodes allows recursive iteration.

Therefore, to access the nodes in question, you could use:

$children = $dom->getElementById("cards")->childNodes
                ->item(1)->childNodes->item(1)->childNodes;

Yet, as you can see this is very messy... Introducing XPath:

Upvotes: 3

Related Questions