Read DIV tag innerHTML

lionelcrottaz

Hi everybody !

I want to put the content of a DIV tag into a variable. Can anyone help ?

<div id="myDiv">blablabla</div>
<?php $content = ?>

Thanks a lot !

Lionel

RossC0

Well really need more information regarding your situtation and how you want to get the div contents:

1) You could use javascript and the innerHTML function - write it to a form and post it to the server - so a php variable can see it.

2) Read the html file - and use a reg expression or sub string function to get the contents...

lionelcrottaz

Hi,

Thanks for your answer. I know how to do it with javacript, but the idea is to do it with php to be usefully for directories submission. So I'd like (2) Read the html file - and use a reg expression or sub string function to get the contents...).

The text I want to extract is contained in a DIV tag which contains the first 300 char of a webpage. I want to put this string into the META description tag to improve ranking. Thanks for your help !

Lionel

RossC0

ok - well its not as easy as it sounds.

Is your html well formed? Could you parse to XML ?

If it is well formed and the div that you are using is straight forward enough - you could use something like:

<?php
$HTML = '<div id="mydiv">HELLO WORLD!</div>';

$html = substr($HTML, strpos($HTML, 'id="mydiv">')+11);
$html = substr($html, strpos($html, 0, '</div>'));

echo $html;

lionelcrottaz

I see what you means. Let me give you part the code description :

index.php contains now :

<?php
function outputDescription()
{
sequence;
}
?>
<html>
<head>
<META name="description" content="<?php outputDescription(); ?>">
</head>
<body>
<div id ="myDiv">
some text jaskd j asdjs daéls jdaféjasdf kljsfékljasfé asjfkd
</div>
</body>
</html>

Now the result I want to obtain :

<html>
<head>
<META name="description" content="some text jaskd j asdjs daéls jdaféjasdf kljsfékljasfé asjfkd">
</head>
<body>
<div id ="myDiv">
some text jaskd j asdjs daéls jdaféjasdf kljsfékljasfé asjfkd
</div>
</body>
</html>

Do you need any other information ? Thanks a lot for your help !

RossC0

I dont think you can dynamically have a function like that which can add the meta tag.

You need to read the contents of the html first - to be able to generate the code to create the meta tag.

So you could
1) write some code to parse the html pages and add the meta tag.
2) Use ob_get_contents to get the html and then insert it.

Drakla

Something like this as the outputdescription should work. I haven't tested it, but it's along the right path, most of this is RossCo's work with a bit tagged on.

function outputDescription()
{
	$html = file_get_contents($_SERVER['SCRIPT_FILENAME']);

// find the starting offset
$startMatch = 'id="mydiv">';
$endMatch = '</div>';

$startOff = strpos($html, $startMatch + strlen($startMatch));
$endOff   = strpos($html, $endMatch, $startOff);

$html = substr($html, $startOff, $endOff - $startOff);

// if you want 300 characters of text AFTER the stripping
// now strip the tags off the stuff inside

$html = strip_tags($html);

echo substr($html, 0, 300);
}

Note you can't have another div inside the main one you're doing the guff for - well you can but it might end prematurely.

lionelcrottaz

Thanks a lot for your help ! It is a very good start !

There will be no sub DIV in myDiv. How can I now exactly put the div extraction content into my META description ?

laserlight

You could consider using regex to match the div element in question.
e.g.

<?php
$str = "<div id=\"myDiv\">hello foo\nbar\tbaz</div>";

if (preg_match('/<div id="myDiv">([\w\s]*)<\/div>/is', $str, $matches)) {
	echo $matches[1];
} else {
	echo 'No match';
}
?>

Basically the regex looks for <div id="myDiv"> in case insensitive format and takes all alphanumeric/underscore/whitespace characters until the first </div> encountered into a subpattern.
$matches[1] then contains this subpattern.

Drakla

Put it in exactly like you did in your example with the

Line

lionelcrottaz

I can see it's partially working, as it outputs only the META title content. It seem that the extracted string is not correct. I try to see what's wrong, but I don't find the reason why. Any idea ?

Drakla

I just tested it and it works fine - only thing I had to change was the capitalisation of mydiv to myDiv and it worked perfectly.

lionelcrottaz

You can check the rusult on : http://www.crottaz.com/new/index.php

The displayed string is only the title content. I don't get it. Code running is :

<?php
function outputDescription()
{
$html = file_get_contents($_SERVER['SCRIPT_FILENAME']);

// find the starting offset 
$startMatch = '<div id="myDiv">';
$endMatch = '</div>'; 

$startOff = strpos($html, $startMatch + strlen($startMatch)); 
$endOff   = strpos($html, $endMatch, $startOff); 

$html = substr($html, $startOff, $endOff - $startOff); 

// if you want 1000 characters of text AFTER the stripping 
// now strip the tags off the stuff inside 

$html = strip_tags($html); 
echo substr($html, 0, 1000);

}
?>

Drakla

Yeah - it's well weird. I'm running the code and it's telling me that the string position of id="myDiv" is 95, which is absolute crap.

I'll modify the code to make it work and slap it up in a sec

And make sure you either put the function in a seperate file and include it, which I assume you're going to do if this is used on loads of pages, OR put the function at the bottom of each page so it doesn't search within itself.

[TIME PASSES] Dammit - had a bracket in the wrong place when calculating the start offset - new and working code, even tested with your page:

function outputDescription()
{
	$html = file_get_contents($_SERVER['SCRIPT_FILENAME']);

// find the starting offset
$startMatch = 'id="myDiv">';
$endMatch = '</div>';

$startOff = strpos($html, $startMatch) + strlen($startMatch);
$endOff   = strpos($html, $endMatch, $startOff);

$html = substr($html, $startOff, $endOff - $startOff);

// if you want 300 characters of text AFTER the stripping
// now strip the tags off the stuff inside

$html = strip_tags($html);

echo substr($html, 0, 300);
}

lionelcrottaz

Hi !

It works perfectly ! Thanks a lot ! I should just put the script at the end of the page and everything is ok !

Thanks a lot !

Lionel