Regexp for html

Question

Possible Duplicate:
RegEx match open tags except XHTML self-contained tags

I have the following string:

$str = " 
<li>r</li>  
<li>a</li>  
<li>n</li>  
<li>d</li>  
...
<li>om</li>  
";

How do I get the HTML for the first n-th <li> tags?

Ex : n = 3 ; result = "<li>r<...>n</li>;

I would like a regexp if possible.

[rant](http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454) — Chris Shouts, Aug 30 '10 at 20:50
All you need to know can be found here: http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454 — prodigitalson, Aug 30 '10 at 20:50
Why all the downvoting? Most newbies *will* make this mistake once (and hopefully only once). Also, OP isn't parsing entire documents, it looks like it's a string of unnested `
prodigitalson - actually you gave me the best answer I could hope for...how was I supposed to find that just by searching ? — johnlemon, Aug 30 '10 at 21:05
@FrustratedWithFormsDesigner - Agreed. This is a valid question that deserves an answer. That being said, I am going home to play with my son :) — Chris Shouts, Aug 30 '10 at 21:06
This question has been asked soooo many times. Are people not searching? Is it not coming up properly in searches? What is going on? It's been asked at least few times just this afternoon! — webbiedave, Aug 30 '10 at 21:10
@webbiedave: OP is not parsing an HTML documents with regex, OP is searching a string with only one type of tag, all at the same level (no nesting - which is usually the main problem with regex + HTML). Not really the same. — FrustratedWithFormsDesigner, Aug 30 '10 at 21:38
@FrustratedWithFormsDesigner: How is there no nesting? His inner html is spelling the word "random" which leads me to believe that anything could be in those tags. — webbiedave, Aug 30 '10 at 22:11
@webbiedavie: It's a very specific sort of example and he never says "any HTMl" or "other tags an' stuff" which leads me to think he will stick to the example (or similar strings); but fair enough: if OP shows that his other input strings can have nested tags, you get 100 internet points and I'll delete my earlier comment. :) — FrustratedWithFormsDesigner, Aug 30 '10 at 22:38

score 10 · Answer 1 · answered Aug 30 '10 at 21:04

10

Like this.

$dom = new DOMDocument();
@$dom->loadHTML($str);
$x = new DOMXPath($dom); 

// we wan the 4th node.
foreach($x->query("//li[4]") as $node) 
{
  echo $node->c14n()
}

Oh yeah, learn xpath, it will save you lots of trouble in the future.

answered Aug 30 '10 at 21:04

Byron Whitlock

52,691
28
123
168

I would always recommend SimpleXML over DOMDocument dor so simple things as DOMDocument needs an overload of additional objects (like for xpath) and makes selecting of elements and their content way to complicated. – 2ndkauboy Aug 30 '10 at 21:11
@Kau-Boy - interesting, care to post an example? Also do realize DOMDocument doesn't require you run the html though tidy. That in itself is a major win for me. – Byron Whitlock Aug 30 '10 at 21:22

2ndkauboy · Answer 2 · 2010-08-30T21:49:32.723

6

The Solution of @Byron but with SimpleXML:

$xml = simplexml_load_string($str);

foreach($xml->xpath("//li[4]") as $node){
  echo $node[0]; // The first element is the text node
}

EDIT: Another reason I really like at simplexml is the easy debugging of the content of a node. You can just use print_r($xml) to print the object with it's child nodes.

edited Aug 30 '10 at 21:49

answered Aug 30 '10 at 21:28

2ndkauboy

9,302
3
31
65

Corrected, thanks for the hint! – 2ndkauboy Aug 30 '10 at 21:42

score 1 · Answer 3 · answered Aug 30 '10 at 21:07

As I'm sure you are aware it is not a good idea to use regular expressions to work through HTML unless you were to "tidy" it first.

A very viable solution in PHP would be to navigate the HTML structure using Simple XML (http://php.net/manual/en/book.simplexml.php) or as a DOM Document (http://php.net/manual/en/class.domdocument.php).

Regexp for html

3 Answers3