Get page source

jdloaner

I need to make a button, and a text field, that when the button is pressed, it will look at the source of the URL in that text field and copy all the text in between the <HEAD> tags of the souce of that URL and display it on the next page.

Why am I having the hardest time with this? Any ideas? Thanks, this is bothering me to the fullest. :mad:

BobLennon

jdloaner wrote:
I need to make a button, and a text field, that when the button is pressed, it will look at the source of the URL in that text field and copy all the text in between the <HEAD> tags of the souce of that URL and display it on the next page.

Why am I having the hardest time with this? Any ideas? Thanks, this is bothering me to the fullest. :mad:

Assuming the URL is in the scope of the php filesystem on the server, you might consider using a regular expression to extract the code within the <head></head> tags.

jdloaner

The URL will be an external website, lets say for instnace google.com.

How can I get the info in between the tags?

michaewlewis

you might consider using an iframe for the google.com part of the page and have the rest of the page reside on your server, that way you can pass a variable in the url and then run an if statement in the next page to decide which content will go in the <head>.

Why do you need to put something in the <head> of the new page?
If you want to make the page use a specific stylesheet, you might need to use a javascript.
Not really sure what you're wanting to accomplish.

jdloaner

All that I am trying to accomplish is to look at an HTML file, find the start of the <HEAD> tag, copy everything in between until </HEAD> then output it.

Ludichrist

Try this

$handle = fopen("http://www.google.ca/index.html",r);
while (!feof($handle)){
	$buffer .= fread($handle,1024);
}

preg_match("/^<head.*>.*<\/head>$/",$buffer,$matches);

echo $matches[0];

jdloaner

That makes sense to me. The array seems to be empty though. It deosnt seem to have a value stored and we obviously know google has <head> tags. Any clue?

<html>
<body>
<?php

$handle = fopen("http://www.google.ca/index.html",r);
while (!feof($handle)){
$buffer .= fread($handle,1024);
}

preg_match("/^{<head.>.<\/head>$/",$buffer,$matches);}

echo $matches[1];

</body>
</html>

stonefish

echo $matches[0];

Ludichrist

Not sure why that regular expression isn't working. Anyway, here's another easy way:

<?
$handle = fopen("http://www.google.ca/index.html",r);
while (!feof($handle)){
	$buffer .= fread($handle,1024);
}
$edited = preg_replace("/</","&lt",$buffer);

$start = strpos($buffer,"<head>");
$end = strpos($buffer,"</head>");

//find what's in between the head tags (Add 7 to account
//for the length of </head>
$head = substr($buffer, $start, ($end - $start) +7);

echo $head;
?>

Oni

Nice code,

I'll add it to my library just in case I need it 🙂

Ludichrist

thanks 🙂

jdloaner

Yes that works much better, thank you. But I am still having a problem capturing all the data between(and including) the tags.

I.E

<html>
<body>
<?
$handle = fopen("http://www.google.com/index.html",r);
while (!feof($handle)){
$buffer .= fread($handle,2048);
}
$start = strpos($buffer,"<head");
$end = strpos($buffer,"/head>");
$head = substr($buffer, $start, $end);
?>

Will return only this in a text box:
<head><meta http-equiv=

Yet the first line of google's source is:
<head><meta http-equiv="content-type" content="text/html; charset=UTF-

jdloaner

Oh yeah, I forgot to add:

$start returns the correct value of '6' but
$end returns '286'

I belive it should be more in the neighborhood of '475'

Is that correct?

jdloaner

Ha.. ok I am an idiot, I got it working.

How is it going to handle multiple sets of tags on the same page?

Ludichrist

It won't handle multiple sets of <head> tags. There shouldn't be more than one set on any html page anyway.

If, for some reason, you want to account for this, you can shorten $head like this:

$head = substr($head,$end+7,strlen($head))

That will give you everything after the closing head tag, and you can use a while loop until there are no more <head> tags.

jdloaner

If there are multiple results I would like to append them onto the end of the first set, but this doesn't seem to be doing it for me.

do{
$head = substr($buffer, $start, ($end - $start) +8);
}
while ($start != false);
break;

Ludichrist

<?
$handle = fopen("http://www.google.ca/index.html",r);
while (!feof($handle)){
    $buffer .= fread($handle,1024);
}

$head = "";

while($strlen($buffer) > 0){
    //leave out the closing < because some head tags have attributes
    $start = strpos($buffer,"<head");
    $end = strpos($buffer,"</head>");

//find what's in between the head tags (Add 7 to account
//for the length of </head>
$head .= substr($buffer, $start, ($end - $start) +7);

$buffer = substr($buffer, $end+7, strlen($buffer));
}

echo $head;
?>

BTW, it's easier for us to read your code if you wrap it in

 tags.

jdloaner

Perfect. I think this should do it. Thanks for your help.

Will you explain to me what the .= does vs =

Thanks.

Weedpacket

Just thought I'd mention:

/^<head.*>.*<\/head>$/

. doesn't match newlines by default (add the /s modifier to get it to do that). matches the start of the string and $ matches the end, so all up, this will match a <head>...</head> only if (a) it's all on one line, and (b) the only thing in the document.

For efficiency, at least the first . could be replaced by .?, save the regexp engine from wasting its time trying to match the > against the very last > in the document before trying to match it against any that appear earlier (. matches > as well).

Ludichrist

jdloaner wrote:
Perfect. I think this should do it. Thanks for your help.

Will you explain to me what the .= does vs =

Thanks.

$string = $another_string does what you'd expect.

$string .= $another_string is short hand for $string = $string . $another_string.

You can use +=, or -=, or *=, etc.. in the same way.