Get the string from a website having a specific word

Question

A webiste www.example.com have many lists in it. That is,

   <ol>
    <li>This is a list saying about asp</li>
    <li>This is a list  saying about javascript</li>
    <li>This is a list saying about php</li>
    <li>This is a list saying about .net</li>
    </ol>

I need to Get the list with a word "php" using php.
That is the output should be "This is a list saying about php"

How can i do this with preg_match???

I used CURL class to fetch the HTML contents. here is the code i used

$site = $curl->get("http://www.example.com/outputs.html");
$pattern = 'I NEED TO GET THIS PATTERN';
preg_match($pattern, $site, $matches);
$php_out = $matches[1];
echo $php_out;

when i use,

$pattern = '/<li>(.*?)<\/li>/s';

It returns the first result
That is "This is a list saying about asp"

possible duplicate of [How to parse and process HTML with PHP?](http://stackoverflow.com/questions/3577641/how-to-parse-and-process-html-with-php) — John Conde, Aug 23 '12 at 12:57
Do you have to fetch the HTML from the page and search for the word or do you already have the HTML in your script? — futuregeek, Aug 23 '12 at 12:59

score 1 · Answer 1 · answered Aug 23 '12 at 15:19

The simplest solution would be to remove the /s modifier. Then each match would be limited to one line, because the . can't match newlines.

But that only works if the HTML is formatted as in your sample, with each element on a separate line and no newlines in the element's content. Here's a more robust solution:

$pattern = '~<li>[^<]*php[^<]*</li>~'

But be aware that there are many things that can cause the regex to fail, even in perfectly valid HTML. Unless this is a one-time, throwaway job you should seriously consider using an HTML-specific tool as the other responders advised.

score 0 · Accepted Answer · answered Aug 23 '12 at 12:59

0

You need a website crawler and a parser. There is a project called PHPCrawl with this lib you can crawl the site and get the content. Then you can parse and search in the source code for the specified pattern. If you want you can do it with a regex.

But i think you are not the first here on Stackoverflow with this problem. Perhaps you should search here and you'll get some more information.

answered Aug 23 '12 at 12:59

René Höhle

26,716
22
73
82

I AM sure that i can solve it using an appropriate regular expression. hope anyone will help me to find it out. – Abhijith P Aug 23 '12 at 13:17
Yes you can do it with a pattern but you need the crawler to follow all sites. – René Höhle Aug 23 '12 at 14:20
why i should follow all sites? I'm getting the result using preg_match() itself. But the problem is that i dont know regular expression. – Abhijith P Aug 23 '12 at 17:29

Get the string from a website having a specific word

2 Answers2