using preg_replace to cull text

d_l

I can use preg_replace to replace "string1" with "string2"

but how can I setup preg_replace (or other regexp function) to delete everything from "string1" to "string2" .. and including the start and end strings?

so if this is in the text to be culled ...

<div id="footer><a href="http://www.w3.org/TR/xhtml11/" xmlns="">XHTML 1.1</a>other info ..</div>

I need to delete from <div id="footer" to </div>

this is my first effort which does not work .. $string is the entire input doc to be parsed

$string = preg_replace( '/(<div id="footer")(\w+)(<\/div>)' , ' ' , $string);

Weedpacket

quote[/quote]This is almost certainly wrong. You'll want (.+?). You also need the closing / at the end of the regexp (to match the opening /), and follow that with s.
/(<div id="footer")(.+?)(<\/div>)/s

d_l

Thanks for solving that problem.

I only needed to escape the quotations inside the regexp expression to get it to work ..

/(<div id=\"footer\")(.+?)(<\/div>)/s

I have a related question ..

what is the opposite of preg_replace?

i.e. to include (non-cull) only the HTML code between string1 and string2

I am trying this ....

preg_match_all( "/(<div id=\"footer\")(.+?)(<\/div>)/s ", $htmlfile, $out, preg_pattern_order);

but I get this error ..

Warning: preg_match_all() expects parameter 4 to be long, string given in

Note: For others learning regexp, this utility is very useful

The Regex Coach

Thanks again.

Weedpacket

Capital letters: PREG_PATTERN_ORDER

d_l

o.k. .. now .. still using preg_match_all .. where/how can I get the number of matches?

I've read that I can inspect the array using

var_dump($matches);

p.s. .. late addition .. answering my question ..

I now see how this works ..

// this expression extracts all contents of table cells .. <td> .. </td>
preg_match_all( "/(<td)(.+?)(<\/td>)/s ", $htmlfile, $matches, PREG_PATTERN_ORDER);

//for testing only, this dumps the array stack
var_dump($matches);

// this echos the number of matches in array $matches
$n = count($matches[0]);
echo $n;

// this echos the values of array

echo "<br />"."matches [0][0] =  ".$matches[0][0];
echo "<br />"."matches [0][1] =  ".$matches[0][1];
echo "<br />"."matches [0][2] =  ".$matches[0][2];
echo "<br />"."matches [0][3] =  ".$matches[0][3];
echo "<br />"."matches [0][4] =  ".$matches[0][4];

// ... through to [0][$n]

And this site explains the PHP5 array stack.

http://www.theopensourcery.com/phparray.htm

Weedpacket

count($matches[0]);

d_l

Thanks .. now working ..

I've found that, in general, for my purpose (scraping HTML content) it is easier to extract the information to be retained - preg_match_all - rather than deleting the information to be excluded - preg_replace