How to retrieve website names? - php

Question

I am wondering if there is a way to get the name of a website from a URL. I know you can parse a URL to get a domain name, but since site names are not standardized as far as code is concerned, I am doubtful.
An example of how this could be used is say I am linking to a New York Times article. I can have the title of the article link to the article page. Then I might want to have the source, "The New York Times" displayed next to the title of the article. It would be exceedingly convenient if I could have this automatically generated.
Just getting the page title wouldn't work because that would usually give you the article title or, if you were to link to some other type of page, you might get some string like "How to retrieve website names? - Stack Overflow." I would only want to get the "Stack Overflow" part of that.
Any ideas?

Answer 1

You could try the application-name property:
<meta name="application-name" content="The New York Times" />
also
<meta name="application-name" content="CNN"/>
Not every site will have this but you can start here, check for open graph tags (http://ogp.me), etc.

Answer 2

You will need to parse the DOM tree using DOMDocument:
<?php
function GetTitle($url)
{
$dom = new DOMDocument;
#$dom->loadHTMLFile($url); // # supresses warnings
// try to get meta application-name
foreach ($dom->getElementsByTagName("meta") as $meta)
{
$metaName = $meta->attributes->getNamedItem("name");
if (strtolower($metaName->nodeValue) == "application-name")
{
$metaContent = $meta->attributes->getNamedItem("content");
if ($metaContent != NULL)
return $metaContent->nodeValue;
}
}
// title fallback:
foreach ($dom->getElementsByTagName("title") as $title)
return $title->nodeValue;
return NULL;
}
print(GetTitle("http://www.nytimes.com/"));
?>
First, GetTitle() looks for a <meta name="application-name"> tag. If not found, it will fallback and return the page title instead.
Additionally, you should pass the base url. F.e. if you have this url: http://stackoverflow.com/questions/16185145/how-to-retrieve-website-names/16185654#16185654, you should strip everything except http://stackoverflow.com using parse_url:
$parsedUrl = parse_url($url);
GetTitle($parsedUrl["scheme"] + "://" + $parsedUrl["host"]);

Answer 3

If you want to parse the url, you could try this function:
$parsedUrl = parse_url($url);
$host = $parsedUrl['host']
echo $host;
This will give you an associative array where the host key is what you are looking for.
See: http://php.net/manual/en/function.parse-url.php

Answer 4

What you call 'Site name' is not a part of the link, it is part of the HTML code returned by that link.
If you want to get the site title, you should retrieve the link content using CURL and then parse the returned HTML to get the content of the tag in the section.
Probably this will be expensier than the benefit you could get.