[RESOLVED] Get HTML Title from a File

[rpanning]

I'm making a system in which scans a file system for valid Web pages and get's a title from the filename. However, an option would be to get the title from the actual HTML <title> in the file. I've been thinking about the most efficient way to do this.

My thoughts at this point is to use SplFileObject and foreach() through each line. Rather than getting the entire contents of the file. I think memory usage would be better that way.

Then I would need to find <title>, start capturing the title after that, then find the closing </title>. One concern is what happens if the two tags are on different lines. So now I'd have to do some detection and trimming to get the correct title.

Any thoughts about how to do this, efficiently? I'm starting at the point of cycling through each line:

foreach ($file as $line) {

}

[scrupul0us]

$doc = new DOMDocument();
$doc->loadHTMLFile($filename);

foreach ($doc->getElementsByTagName('title') as $nodes)
{
     $pageTitle[] = $nodes->nodeValue;
}

[rpanning]

@ scrupul0us: That is a option, but it opens the entire file. I'm trying to keep it efficient and not use so much memory.

I have worked through the issue already and finished the method. Here is what I put together:

	public function getHtmlTitle()
	{
		// Open the file
		try {
			$file = $this->openFile();
		} catch (Exception $exception) {
			return false;
		}

	// Iterate over each line
	foreach ($file as $line) {

		// Find the tags
		$pos_open  = stripos($line, '<title>');
		$pos_close = strripos($line, '</title>');

		// Open and close on the same line
		if ($pos_open !== false && $pos_close !== false) {
			$start  = $pos_open + 7;
			$length = $pos_close - $start;
			$title  = substr($line, $start, $length);
			break;
		}

		// Open one one line
		if ($pos_open !== false) {
			$start = $pos_open + 7;
			$title = substr($line, $start);
			continue;
		}

		// Close on one line
		if ($pos_close !== false && isset($title)) {
			$title .= substr($line, 0, $pos_close);
			break;
		}

		// Continue title if started
		if (isset($title)) {
			$title .= $line;
		}
	}

	// Title not found
	if (!isset($title)) {
		return false;
	}

	// Cleanup and return
	$title = trim($title);
	$title = str_replace(array("\r\n", "\n\r", "\r", "\n"), ' ', $title);
	return $title;
}

[Weedpacket]

            $pos_open  = stripos($line, '<title>'); 

So there's no chance that any of the files you're looking at will have title elements with attributes, or whitespace before the '>'?

[rpanning]

Weedpacket;10904797 wrote:

So there's no chance that any of the files you're looking at will have title elements with attributes, or whitespace before the '>'?

I was trying to look into that. I never have had an attribute in the <title>. But I couldn't find anything that said there was a valid attribute. Do you know if they are allowed? If so, what would be a good solution?

[Weedpacket]

Sure they are. Just have a look at what the HTML specification says about it.

You can look for "<title" and "</title" instead.

[rpanning]

Weedpacket;10904827 wrote:

Sure they are. Just have a look at what the HTML specification says about it.

You can look for "<title" and "</title" instead.

I did take a quick glance at the spec but didn't find anything that directly said attributes were allowed. The only ones I would think of are class and id but those are mute.

Then I need to find the closing > so I know where to start capturing the title. I'll work on this.

[Weedpacket]

Section &#167;7.4.2 of the HTML 4.0.1 Specification, "The TITLE Element".

[rpanning]

Weedpacket;10904877 wrote:

Section §7.4.2 of the HTML 4.0.1 Specification, "The TITLE Element".

That's what I was looking for, must have missed the attribute info. Anyway, here is what I've come up with, basically 2 changes:

$pos_open = stripos($line, '<title');
..and..
$start = strpos($line, '>', $pos_open) + 1;

	public function getHtmlTitle()
	{
		// Open the file
		try {
			$file = $this->openFile();
		} catch (Exception $exception) {
			return false;
		}

	// Iterate over each line
	foreach ($file as $line) {

		// Find the tags
		$pos_open  = stripos($line, '<title');
		$pos_close = strripos($line, '</title>');

		// Open and close on the same line
		if ($pos_open !== false && $pos_close !== false) {
			$start  = strpos($line, '>', $pos_open) + 1;
			$length = $pos_close - $start;
			$title  = substr($line, $start, $length);
			break;
		}

		// Open one one line
		if ($pos_open !== false) {
			$start = strpos($line, '>', $pos_open) + 1;
			$title = substr($line, $start);
			continue;
		}

		// Close on one line
		if ($pos_close !== false && isset($title)) {
			$title .= substr($line, 0, $pos_close);
			break;
		}

		// Continue title if started
		if (isset($title)) {
			$title .= $line;
		}
	}

	// Title not found
	if (!isset($title)) {
		return false;
	}

	// Cleanup and return
	$title = trim($title);
	$title = str_replace(array("\r\n", "\n\r", "\r", "\n"), ' ', $title);
	return $title;
}