Unalix

Unalix is a dependency-free Python library for removing tracking fields from URLs, following the ClearURLs specification.

This is an actively maintained fork of AmanoTeam/Unalix, which has been unmaintained since February 2022. This fork adds extended tracking parameter coverage, a custom rules API, IDN homograph attack detection, SSRF protections, and automated ruleset updates.

What's new in this fork

• Extended ruleset (unalix_extended.json) -- 25 additional tracking parameters (Google Ads, Microsoft, TikTok, LinkedIn, HubSpot, MailChimp, etc.) and 7 redirect services (Slack, Skimlinks, CJ Affiliate, etc.) not covered by ClearURLs
• Custom rules API -- rulesets_from_dict() and custom_rulesets parameter on clear_url() for user-defined rulesets
• Homograph detection -- detect_homograph() identifies IDN homograph attacks using Cyrillic/Greek/fullwidth lookalike characters
• Security hardening -- SSRF protection in unshort_url(), capped Retry-After sleep, fixed IPv6 parsing, forward-secret-only TLS ciphers, proper cookie policy subclasses
• Automated updates -- GitHub Actions for weekly ClearURLs ruleset and CA bundle updates via PR
• Updated data -- ClearURLs rules updated from 173 to 206 providers, CA bundle updated to 2026
• Comprehensive tests -- 116 tests (up from 2), covering real-world URLs from clean-url, url-tracking-stripper, Link Cleaner+, and link-cleaner-pro

Installation

Install from this fork:

pip install 'https://codeload.github.com/DigitalCyberSoft/Unalix/tar.gz/refs/heads/master'

Note: Unalix requires Python 3.8 or higher.

Usage

Removing tracking fields

import unalix

url = "https://example.com/page?utm_source=google&fbclid=abc&gclid=xyz"
result = unalix.clear_url(url)

assert result == "https://example.com/page"

Resolving shortened URLs

import unalix

url = "https://bitly.is/Pricing-Pop-Up"
result = unalix.unshort_url(url)

assert result == "https://bitly.com/pages/pricing"

Tip: unshort_url() strips tracking fields from every URL before following a redirect, so you don't need to call clear_url() separately on its return value.

Custom rules

import unalix

custom = unalix.rulesets_from_dict({
    "providers": {
        "myapp": {
            "urlPattern": ".*",
            "rules": ["my_tracking_param"]
        }
    }
})

url = "https://example.com/?my_tracking_param=abc&page=1"
result = unalix.clear_url(url, custom_rulesets=custom)

assert result == "https://example.com/?page=1"

Homograph attack detection

import unalix

# Cyrillic 'о' (U+043E) masquerading as Latin 'o'
result = unalix.detect_homograph("https://g\u043e\u043egle.com")

assert result is not None
assert result["normalized"] == "google.com"
assert result["target"] == "google.com"

Contributing

If you have discovered a bug in this library and know how to fix it, fork this repository and open a Pull Request.

Third party software

Unalix includes some third party software in its codebase. See them below:

• ClearURLs

  • Author: Kevin Röbert
  • Repository: ClearURLs/Rules
  • License: GNU Lesser General Public License v3.0

• Requests

  • Author: Kenneth Reitz
  • Repository: psf/requests
  • License: Apache v2.0

• Pyrogram

  • Author: Dan
  • Repository: pyrogram/pyrogram
  • License: LGPL-3.0