0

Possible Duplicates:
Xpath not behaving for me in parsing basic html

I know how to get content from a div with static name (i.e. always the same in the whole page). However, my case is "post_id_xxxxx", something like this:

<div id="post_id_12345">abc</div>

<div id="post_id_67890">abc</div>

<div id="post_id_31234">abc</div>

I would like to extract the "abc" string, but seems difficult to me since every div has different ID.

Thanks.

Community
  • 1
  • 1
Matt
  • 1
  • 1
    You don't. [Period](http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454). –  Mar 13 '11 at 22:03
  • possible duplicate of [Xpath not behaving for me in parsing basic html](http://stackoverflow.com/questions/5249492/xpath-not-behaving-for-me-in-parsing-basic-html). Mind the comments to the Accepted Answer. You should be able to find plenty of additional duplicates if you just bother to use the search function. – Gordon Mar 13 '11 at 22:13

2 Answers2

1

Do not parse HTML/XML with regexp. HTML has a structure that a html specific parser can exploit. See this classic link: RegEx match open tags except XHTML self-contained tags

You should try some of PHPs parsers like domdocument

DO NOT USE THIS

Here is a regexp that will match the example you specified. It will not work on more complicated structures (e.g. nested divs). You haven't really specified what invariants you know about the structure of your html, from the example this should work. You can expand this regexp to match more complexities, but a real parser will be much more robust and easier.

<div id="post_id_[0-9]{5}">(.*)</div>
Community
  • 1
  • 1
Jesse Cohen
  • 4,010
  • 22
  • 25
  • About DOMDocument, I tried to use it but it throws errors every time I want to use it on a HTML document, like this: Warning: DOMDocument::loadXML() [domdocument.loadxml]: Entity 'aacute' not defined in Entity, line: 35 in test.php on line 10 Warning: DOMDocument::loadXML() [domdocument.loadxml]: Opening and ending tag mismatch: SCRIPT line 48 and script in Entity, line: 66 in test.php on line 10 – Matt Mar 13 '11 at 22:17
  • @Matt because you are using it wrong. Use `loadHTML` and `libxml_use_internal_errors`. Just search SO for DOM and PHP. All your UseCases have been answered before. – Gordon Mar 13 '11 at 22:20
  • don't use loadXML use loadHTML, the loadHTML method should be relatively robust to malformed html and should be able to correct it into a dom tree that you can parse (usually). – Jesse Cohen Mar 13 '11 at 22:21
1

This is still workable with a regex, if it's really only about the overly simplistic cases in your example:

preg_match('#<div\s[^>]*id="post_id_12345"[^>]*>(.*?)</div>#', $str, $m)

But as soon as you have nested divs in the document or other complex constructs, you need to use a HTML parser. To give you a real example instead of generic links, use phpQuery or QueryPath with:

print qp($html)->find("#post_id_12345")->text();
mario
  • 144,265
  • 20
  • 237
  • 291