From 3cf51dd8744d98848b31dee0e095d6167a61ca40 Mon Sep 17 00:00:00 2001
From: msramalho <19508417+msramalho@users.noreply.github.com>
Date: Wed, 11 Jun 2025 11:56:42 +0100
Subject: [PATCH] adds tracker remove feature and tests

---
 src/auto_archiver/core/orchestrator.py |  4 +--
 src/auto_archiver/utils/url.py         |  8 ++++--
 tests/utils/test_urls.py               | 37 ++++++++++++++++++++++++++
 3 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/src/auto_archiver/core/orchestrator.py b/src/auto_archiver/core/orchestrator.py
index 0bff376..66073a7 100644
--- a/src/auto_archiver/core/orchestrator.py
+++ b/src/auto_archiver/core/orchestrator.py
@@ -34,7 +34,7 @@ from .config import (
 from .module import ModuleFactory, LazyBaseModule
 from . import validators, Feeder, Extractor, Database, Storage, Formatter, Enricher
 from .consts import MODULE_TYPES, SetupError
-from auto_archiver.utils.url import check_url_or_raise
+from auto_archiver.utils.url import check_url_or_raise, clean
 
 if TYPE_CHECKING:
     from .base_module import BaseModule
@@ -572,7 +572,7 @@ Here's how that would look: \n\nsteps:\n  extractors:\n  - [your_extractor_name_
             raise e
 
         # 1 - sanitize - each archiver is responsible for cleaning/expanding its own URLs
-        url = original_url
+        url = clean(original_url)
         for a in self.extractors:
             url = a.sanitize_url(url)
 
diff --git a/src/auto_archiver/utils/url.py b/src/auto_archiver/utils/url.py
index 2bb19cf..a44a91d 100644
--- a/src/auto_archiver/utils/url.py
+++ b/src/auto_archiver/utils/url.py
@@ -1,5 +1,5 @@
 import re
-from urllib.parse import urlparse, urlunparse
+from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse
 from ipaddress import ip_address
 
 
@@ -53,7 +53,11 @@ def domain_for_url(url: str) -> str:
 
 
 def clean(url: str) -> str:
-    return url
+    TRACKERS = {"utm_source", "utm_medium", "utm_campaign", "utm_term", "utm_content", "fbclid", "gclid"}
+
+    parsed = urlparse(url)
+    clean_qs = [(k, v) for k, v in parse_qsl(parsed.query) if k not in TRACKERS]
+    return parsed._replace(query=urlencode(clean_qs)).geturl()
 
 
 def is_auth_wall(url: str) -> bool:
diff --git a/tests/utils/test_urls.py b/tests/utils/test_urls.py
index df8e0f3..2c77122 100644
--- a/tests/utils/test_urls.py
+++ b/tests/utils/test_urls.py
@@ -1,5 +1,6 @@
 import pytest
 from auto_archiver.utils.url import (
+    clean,
     is_auth_wall,
     check_url_or_raise,
     domain_for_url,
@@ -158,3 +159,39 @@ def test_twitter_best_quality_url(url, best_quality):
 )
 def test_get_media_url_best_quality(input_url, expected_url):
     assert get_media_url_best_quality(input_url) == expected_url
+
+
+@pytest.mark.parametrize(
+    "input_url,expected_url",
+    [
+        # No trackers present
+        ("https://example.com/page?foo=bar&baz=qux", "https://example.com/page?foo=bar&baz=qux"),
+        # Single tracker present
+        ("https://example.com/page?utm_source=google&foo=bar", "https://example.com/page?foo=bar"),
+        # Multiple trackers present
+        ("https://example.com/page?utm_source=google&utm_medium=email&utm_campaign=spring", "https://example.com/page"),
+        # Trackers mixed with other params
+        (
+            "https://example.com/page?foo=bar&utm_content=abc&baz=qux&gclid=123",
+            "https://example.com/page?foo=bar&baz=qux",
+        ),
+        # Only trackers present
+        ("https://example.com/page?utm_source=google&gclid=123", "https://example.com/page"),
+        # No query string
+        ("https://example.com/page", "https://example.com/page"),
+        # Trackers in fragment (should not be removed)
+        ("https://example.com/page#utm_source=google", "https://example.com/page#utm_source=google"),
+        # Trackers after fragment
+        ("https://example.com/page?utm_source=google#section-1", "https://example.com/page#section-1"),
+        # Trackers with empty value
+        ("https://example.com/page?utm_source=&foo=bar", "https://example.com/page?foo=bar"),
+        # Trackers with multiple values
+        ("https://example.com/page?utm_source=google&utm_source=bing&foo=bar", "https://example.com/page?foo=bar"),
+        # Trackers with encoded values
+        ("https://example.com/page?utm_source=google%20ads&foo=bar", "https://example.com/page?foo=bar"),
+        # Unrelated param with similar name
+        ("https://example.com/page?utm_sourc=keepme&foo=bar", "https://example.com/page?utm_sourc=keepme&foo=bar"),
+    ],
+)
+def test_clean_removes_trackers(input_url, expected_url):
+    assert clean(input_url) == expected_url