Grab certain line/place of html file?

Is there a way to grab a html file from another server, then 3 specific words from the file?

These words are static.

I know the first part can be done via curl, which i have installed, but how would i go about it?

Thanks.

Edit: An example is, that i want the result to be that the line i want is put into a variable.

E.g. I want the script to grab line 42, Col 32-40 from http://news.bbc.co.uk/1/hi/world/middle_east/7996962.stm and put width=974 into $grabbed.

 

hmm. What are you looking to do? You just want content from the page or the coding of the page?

You can use something like this:

<?php
/**
 * Simple cURL Script by Dennis M.
 *
 */

// Begin cURL session
$ch = curl_init();

// Define options
curl_setopt($ch, CURLOPT_URL, "LINKTO/PAGE.PHP"); // The URL to go to
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1); // Set to 1 if you want page to display with print

// Execute
$data = curl_exec($ch);

/**
 * From here you can work with the data you've received
 * from the server. If you want to match certain words
 * everytime you can use preg_match and regexp. If you
 * just simply wish to print the page use the print function
 * if there's something else, let me know on the forum! :)
 *
 */

// Let's just see the page
print $data;

// Cleanup..
curl_close($ch);

?>

PHP:

Regards,
Dennis M.

 

Thanks for the reply Dennis.

I want to grab a certain line from the page's source code, which i have no idea how to go about.

The actual line number is static, but a part of the line is dynamic to the url and that is what i want to grab. However, if grabbing 3-4 lines of the code is much easier, i could just do that.

 

Is part of the line you want to grab static? As in, can you be sure the same text and/or tags appear around the dynamic text you want?

If so, use a regular expression to get what you want into a variable.

For example... using Dennis's code above and assuming is your dynamic test was always surrounded like so...

The answer is [dynamic text here]!

You could say something like...

preg_match('/The answer is (.+?)!/', $data, $matches);
echo $matches[0];

 

To grab line 42, you just need this (no need to mess around with curl):

$bbc = file("http://news.bbc.co.uk/1/hi/world/middle_east/7996962.stm");
$line = $bbc[41];

Code (markup):

But if what you want is that 'width' value, I'd do it this way:

$bbc = file_get_contents("http://news.bbc.co.uk/1/hi/world/middle_east/7996962.stm");
if (preg_match('`<meta name="viewport" content="(width=\d+)"/>`', $bbc, $matches))
{
   $width_str = $matches[1];
   echo "Got it: {$width_str}";
}

Code (markup):

It's more robust than expecting the tag to always be on exactly the same line; at any time the BBC may add or remove headers or tags that push it up or down in the source.

 

Thanks guys, perfect!