PHP parse with DOM (no results)

Question

I am trying to retreive the body text located in this span class attribute.

<span id="" style="color:#525B64;">The quick brown fox jumped over the lazy dog.</span>

I tested it on my web server and I get no errors but the page is blank. I'm very new to this so I do not know where to go from here.

Here is my code.

<?php
// Load remote file, supress parse errors
libxml_use_internal_errors(TRUE);
$dom = new DOMDocument;
$dom->loadHTMLFile('http://somewebpage.com');
libxml_clear_errors();

// use XPath to find all nodes with a class attribute of header
$xp = new DOMXpath($dom);
$nodes = $xp->query('//span[@class="msgBody"]');

// output first item's content
echo $nodes->item(0)->nodeValue;
?>

What's the actual page you're trying to access? As far as I can tell, this code works. — lonesomeday, Feb 02 '11 at 01:24

score 3 · Answer 1 · answered Feb 02 '11 at 01:24

3

Everything seems fine in this code.

What I'd try to do is:

remove the line which supresses parse errors.
load the remote file with file_get_contents to see if it loads properly
query document with XPath like //* and loop over resulting DOMNodeList (with foreach) to see if the tree is built correctly.

Btw. to surpress parse errors reported by ->loadHTMLFile() method I use @ operator.

answered Feb 02 '11 at 01:24

mateusza

5,341
2
24
20

1

Personally, to suppress parse errors reported by loadHTMLFile(), I fix the errors. ;-) – Wiseguy Feb 02 '11 at 01:27
Well, that's a fair statement. Thanks for raining on my parade. – Wiseguy Feb 02 '11 at 01:30
BTW, +1 for considering that `allow_url_fopen` may be turned off, disallowing the code from loading a remote file URL. – Wiseguy Feb 02 '11 at 01:34

score 1 · Answer 2 · answered Feb 02 '11 at 01:24

The DOM creates nodes for everthing: attributes, text, comments, elements, you name it. So you're not after the value of the span node even though it might seem that way, you actually want to get the TextNode inside of the span and get its value instead. Try something like:

echo $nodes->item(0)->childNodes->item(0)->nodeValue

You can also get this directly from the xpath query:

$nodes = $xp->query('//span[@class="msgBody"]/text()');

(Though I've never had much luck with xpath, personally.)

I tried this and got the following error "Fatal error: Call to a member function item() on a non-object" — Dan, Feb 02 '11 at 05:52

score 1 · Answer 3 · answered Feb 02 '11 at 01:26

1

Are you sure there is only one span element with this class in the document you are parsing?

Maybe ->item(0) returns empty element and the desired element is next on the list?

answered Feb 02 '11 at 01:26

mateusza

5,341
2
24
20

<-- This span is empty so the page is blank! Everything is correct! – mateusza Feb 02 '11 at 01:43

score 1 · Answer 4 · answered Feb 02 '11 at 04:09

1

Very often such behavior is due to a default namespace (check to see if there is something similar to this: xmlhs="http://www.w3.org/1999/xhtml").

Using in XPath expressions element names that are in default namespace, is the most FAQ in the xpath tag -- just search for "xpath default namespace" to find many good answers.

answered Feb 02 '11 at 04:09

Dimitre Novatchev

240,661
26
293
431

@PeterSmith yes there is this – Dan Feb 02 '11 at 05:45
@Dan: For example how to register namespaces in PHP (this solves the default namespace issue) see this: http://stackoverflow.com/questions/3864359/using-xpath-on-a-php-simplexml-object-soap-namespaces-not-working – Dimitre Novatchev Feb 02 '11 at 06:00

PHP parse with DOM (no results)

4 Answers4