Master URL Analysis: Complete Guide to Regex Patterns & Domain Statistics
How to pull URLs out of messy text, filter them with regex, and read the domain breakdown — with the patterns worth keeping.
Like it ? share it
A migration export, a log file, a page of HTML, a spreadsheet column somebody pasted into an email. URLs arrive surrounded by other text, and the first thing you usually want is a clean list of just the ones that match some pattern.
The URL Analyzer does that in three steps: pull the URLs out, filter them with a regex, and break down what is left.
Getting the URLs out
Paste whatever you have. The tool finds URLs inside plain text, HTML source, JSON, CSV and log output, so you do not need to clean the input first. Mixed content is fine — a paragraph with three links in it produces three URLs.
It reports total URLs found and unique URLs separately, and the gap between those two numbers is often the first useful thing you learn.
Filtering with regex
This is where a list of three thousand becomes a list of forty.
| What you want | Pattern |
|---|---|
| Blog posts | /blog/ |
| Any content section | /(blog|news|articles)/ |
| Versioned API endpoints | /api/v[0-9]/ |
| A specific domain | example\.com |
| HTTPS only | ^https:// |
| Anything with query parameters | \?.+= |
| Subdomains | https?://[^/]+\.[^/]+\. |
| WordPress paths | /wp-admin/|/wp-content/ |
Two things worth knowing when you write your own. A dot matches any character unless you escape it, so example.com also matches examplexcom, and example\.com is what you meant. And the tool validates the pattern as you type, so a broken expression tells you rather than silently matching nothing.
Reading the domain breakdown
The domain statistics answer questions a flat list cannot.
Domain type splits generic TLDs (.com, .org, .net) from country-code ones (.uk, .de, .cn). On an outbound link audit this tells you the geographic shape of who you are linking to.
Domain level counts the labels. example.com is two, blog.example.com is three, api.v1.example.com is four. A migration export full of level-three and level-four hosts means subdomains you may have forgotten exist.
Top domains ranks what appears most. On a competitor's outbound links, this is the whole answer to who they cite.
Each figure is clickable, and clicking through gives you the URLs behind it rather than just the count.
What it is actually for
Before a migration. Filter to /(blog|news|articles)/ and you have the list of content URLs that need redirects, separated from everything else in the export.
Auditing outbound links. Extract every URL from a set of pages and look at top domains. You find out who you are sending readers to, which is usually not the list you would have guessed.
Cataloguing an API. /api/v[0-9]/ against a documentation dump gives you every versioned endpoint, including the old versions somebody forgot to remove.
Security spot-check. /wp-admin/|/wp-content/ against a crawl shows you WordPress paths that are reachable when they should not be.
Where it runs
In your browser. The extraction, the regex and the statistics all happen on your machine, so a migration export full of unreleased URLs does not leave it.