Parsing only part of a document with expat in PHP

Question

I am building a website in PHP in which the content is stored in an XML file. Basically I have a single index.php page, which checks the querystring and serves the appropriate page from the XML.

For example, entering www.mysite.com/?page=home would cause the PHP script to check the XML file for a <page id="home"> tag and paste whatever is inside that tag into index.php.

The contents of <page> tags are stored as HTML, thus:

<xmlroot>
  <page id="home">
    <h1>An HTML Header Tag!</h1>
    <p>This is a paragraph</p>
  </page>
  [...etc]
</xmlroot>

I was hoping to be able to grab the appropriate <page> tag and somehow parse the contents. I know that everything in the <page> tag is valid HTML, so I was just going to use expat to run through the tags and echo them straight back out.

So I am using the DOMDocument method to find the correct <page>, which works fine, except that the contents are returned as a DOM element. The expat parser requires a string. So I need to do one of two things:

Magically convert the DOM element to a string that keeps all the tags intact so I can use it in the expat parser. However, if I could do this I wouldn't need the expat parser, I could just echo that converted string straight out....
Use something other than expat.

Incidentally, I know I could just replace the < and > in the XML with < and >, but this makes the code quite hard to read and edit. I'd like to avoid it if possible.

Damn, I just remembered about CDATA... I'm gonna give it a try. I'll update this if I manage to do it. — David John Welsh, Dec 14 '11 at 08:56
Okay well using CDATA solved it. I'll post the full answer to help anyone else who had the same problem. It'll have to wait till stackoverflow lets me, though, which will be in about 7 hours... :-) — David John Welsh, Dec 14 '11 at 09:05

score 0 · Accepted Answer · answered Dec 14 '11 at 09:12

0

<?php

$doc = new DOMDocument('1.0');

$root = $doc->createElement('html');
$root = $doc->appendChild($root);

$head = $doc->createElement('head');
$head = $root->appendChild($head);

$title = $doc->createElement('title');
$title = $head->appendChild($title);

$text = $doc->createTextNode('< This is the title >');
$text = $title->appendChild($text);

echo $head->ownerDocument->saveXML($head);

DOMDocument::saveXML() takes $node parameter to output only a specific node

http://www.php.net/manual/en/domdocument.savexml.php

answered Dec 14 '11 at 09:12

kapranov.anton

423
5
14

I did take a look at saveXML before posting this question, but it didn't quite work for what I was doing. Thanks anyway, though! – David John Welsh Dec 16 '11 at 01:52
Oh, I just realised what you meant wasn't what I thought you did. I removed the CDATA tags and parsed it using saveXML and it worked great! – David John Welsh Dec 17 '11 at 13:36

score 0 · Answer 2 · answered Dec 15 '11 at 09:02

D'oh. If only I'd waited a few more minutes before asking the question.

The solution was to just encase everything in CDATA tags thus:

<xmlroot> 
  <page id="home"> 
    <![CDATA[
      <h1>An HTML Header Tag!</h1> 
      <p>This is a paragraph</p> 
    ]]>
  </page>
  [...etc]
</xmlroot>

I can then grab the page tag and output the contents using $page->nodeValue. I offer this in the hopes that it saves someone else the pointless headache I just gave myself.

Parsing only part of a document with expat in PHP

2 Answers2