CODEWITHSUNDEEP
phpregexexpressionhtml-parsing
omar dealo
omar dealo

Reputation: 93

Regular expression pattern to match only links without www

I'm trying to search for only links without www like http://google.com, OR https://facebook.com, etc. Then I want to add www to same link so it becomes http://www.google.com, OR https://www.facebook.com, etc.

However, I have a problem in my pattern (the pattern I used to get all links with or without www).

$text = '<a href="http://google.com">google</a> bla bla bla <a href="https://www.google.com">google</a>';
preg_match_all("/<a\s[^>]*href=(\"??)([^\" >]*?)\\1[^>]*>(.*)<\/a>/siU", $text, $matches);
foreach ($matches[2] as $old_url) 
{
$text = str_replace("$old_url","$new_url",$text);
}

Upvotes: 1

Views: 152

Answers (2)

Wiktor Stribiżew
Wiktor Stribiżew

Reputation: 626794

Here is the sample code with the <a\\s[^>]*href=([\"']?)(?>https?|ftps?):\/\/(?![^'\">]*www[^\"]+?\\1)([^'\">]+?)\\1[^>]*>(.*?)<\\/a> regex to match only those URLs in href attributes that have no www in them.

Sample code:

$re = "/<a\\s[^>]*href=([\"']?)(?>https?|ftps?):\/\/(?![^'\">]*www[^\"]+?\\1)([^'\">]+?)\\1[^>]*>(.*?)<\\/a>/"; 
$str = "<a href=\"http://google.com\">google</a> bla bla bla <a href=\"https://www.google.com\">google</a> bla bla bla <a href=\"http://facebook.com\">facebook</a>\n"; 
print ($str . "\n");
$str = preg_replace_callback(
    $re,
    function ($matches) {
        return str_replace($matches[2], "www." . $matches[2], $matches[0]);
  },
  $str
);
print ($str);

Output:

<a href="http://www.google.com">google</a> bla bla bla <a href="https://www.google.com">google</a> bla bla bla <a href="http://www.facebook.com">facebook</a>

Upvotes: 0

hwnd
hwnd

Reputation: 70732

I would consider using DOM and XPath to take care of this for you.

$doc = new DOMDocument;
@$doc->loadHTML($html);

$xpath = new DOMXPath($doc);
$links = $xpath->query('//a[not(contains(@href, "www."))]/@href');

foreach ($links as $link) {
   // process yours urls by $link->nodeValue
   ...
   ...
 }

You could probably then use parse_url() to replace while processing the url.

Upvotes: 2

Related Questions