PHP中使用preg_match提取HTML文档lang属性值的正则匹配优化需求

Got it, let's fix those regex issues you're facing! Your current pattern has two main problems: it expects lang to be the first attribute in the <html> tag, and it doesn't properly limit the capture to just the language code when other attributes follow. Here are two solid solutions:

Solution 1: Improved Regular Expression #

This regex will handle lang attributes anywhere in the <html> tag, and only capture valid language code formats (letters and hyphens, like en, de-DE, etc.):

$sitelang = preg_match('!<html\b[^>]*lang=&quot;([a-z-]+)&quot;[^>]*>!i', $result, $matches) 
    ? $matches[1] 
    : 'Site Language not detected';

Breakdown of the regex: #

• !<html\b : Ensures we match the full <html> tag (the \b word boundary prevents matches for tags like <html5>)
• [^>]* : Matches any characters except > (covers all other attributes or whitespace before/after the lang attribute)
• lang="([a-z-]+)" : Captures only valid language code characters (letters and hyphens) inside the escaped quotes. Using [a-z-]+ instead of .*? avoids accidentally grabbing extra content from subsequent attributes.
• [^>]*> : Matches any remaining content in the <html> tag until the closing >
• !i : Case-insensitive modifier, so it works with <HTML> or LANG=

Test cases it handles: #

• <html lang=&quot;de&quot;> → Returns de
• <html lang=&quot;en&quot; class=&quot;desktop-view not-mobile-device text-size-normal anon&quot;> → Returns en
• <html data-n-head-ssr lang=&quot;en&quot;> → Returns en
• <HTML LANG=&quot;en-GB&quot; data-test=&quot;123&quot;> → Returns en-GB

Solution 2: Use DOMDocument (More Reliable for HTML) #

Regex can struggle with edge cases like malformed HTML, line breaks in tags, or unusual attribute formatting. For a more robust approach, use PHP's built-in DOM parser:

// Initialize DOMDocument and suppress parsing errors (common for real-world HTML)
$dom = new DOMDocument();
libxml_use_internal_errors(true);
$dom->loadHTML($result);
libxml_clear_errors();

// Get the <html> tag and extract the lang attribute
$htmlTag = $dom->getElementsByTagName('html')->item(0);
$sitelang = $htmlTag ? trim($htmlTag->getAttribute('lang')) : 'Site Language not detected';

This method doesn't care about attribute order or tag formatting—it directly accesses the lang attribute of the <html> element, which is the proper way to parse HTML content.

内容的提问来源于stack exchange，提问作者mem