求匹配非http/https前缀的网站域名的正则表达式

Got it, let's break down how to build a regex that targets domain names not prefixed with http:// or https://—even when they're tucked into longer strings like HTML markup or plain text.

The Core Regex #

Here's a solid starting point that works in most modern regex engines (Python, Java, JavaScript ES2018+, etc.):

(?<!https?:\/\/)(www\.[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+)

Let's Break It Down #

• (?<!https?:\/\/): This is a negative lookbehind assertion. It ensures that immediately before the domain we're matching, there isn't http:// or https:// (the s? makes the "s" optional to cover both HTTP and HTTPS).
• www\.: Matches the standard www. prefix (we'll cover adjusting this for non-www domains below).
• [a-zA-Z0-9-]+: Matches the main domain name (letters, numbers, and hyphens are all valid here per domain naming rules).
• \.[a-zA-Z0-9-.]+: Matches the top-level domain (TLD) and any nested sub-TLDs (like .co.in or .io). The hyphen at the end accounts for rare TLDs with hyphens, such as .museum.

Example Matches & Non-Matches #

What gets matched:

• www.google.co.in
• <p>Check out www.w3schools.com for free tutorials</p>
• My go-to resource is www.abc.com

What doesn't get matched:

• http://www.google.co.in
• https://www.w3schools.com/html/
• <a href="https://example.com">Visit our official site</a> (the href value is ignored since it has the HTTPS prefix)

Adjustments for Edge Cases #

If you need to match domains that don't start with www (like example.com), use this modified regex:

(?<!https?:\/\/)([a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+)

Just keep in mind this might accidentally match non-domain strings like file.txt or user@domain.com—so if you're strictly targeting website domains, sticking with the www. version is safer unless you add extra checks (like limiting TLD length to 2-6 characters).

Notes on Regex Engine Support #

• For older JavaScript environments (pre-ES2018), negative lookbehinds aren't supported. In that case, you can first match all potential domains and then filter out any that start with http:// or https:// using basic string methods.

内容的提问来源于stack exchange，提问作者biff