From c4a4eba4a31c43d71bbf4c64ae712d230b4fd001 Mon Sep 17 00:00:00 2001
From: Will Kahn-Greene <willkg@mozilla.com>
Date: Fri, 6 Oct 2023 14:08:18 -0400
Subject: [PATCH] Fix linkify with arrays in querystring (#436)

---
 bleach/linkifier.py   | 4 ++--
 tests/test_linkify.py | 8 ++++++++
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/bleach/linkifier.py b/bleach/linkifier.py
index 54a432e1..8fcefb2c 100644
--- a/bleach/linkifier.py
+++ b/bleach/linkifier.py
@@ -45,8 +45,8 @@ def build_url_re(tlds=TLDS, protocols=html5lib_shim.allowed_protocols):
         r"""\(*  # Match any opening parentheses.
         \b(?<![@.])(?:(?:{0}):/{{0,3}}(?:(?:\w+:)?\w+@)?)?  # http://
         ([\w-]+\.)+(?:{1})(?:\:[0-9]+)?(?!\.\w)\b   # xx.yy.tld(:##)?
-        (?:[/?][^\s\{{\}}\|\\\^\[\]`<>"]*)?
-            # /path/zz (excluding "unsafe" chars from RFC 1738,
+        (?:[/?][^\s\{{\}}\|\\\^`<>"]*)?
+            # /path/zz (excluding "unsafe" chars from RFC 3986,
             # except for # and ~, which happen in practice)
         """.format(
             "|".join(sorted(protocols)), "|".join(sorted(tlds))
diff --git a/tests/test_linkify.py b/tests/test_linkify.py
index 4cc01a6f..0920c470 100644
--- a/tests/test_linkify.py
+++ b/tests/test_linkify.py
@@ -300,6 +300,14 @@ def test_link_ftp():
     )
 
 
+def test_link_with_qs_with_array():
+    """Test that urls pick up [] in querystring"""
+    assert linkify("http://test.com?array[]=1&params_in[]=2") == (
+        '<a href="http://test.com?array[]=1&amp;params_in[]=2" '
+        + 'rel="nofollow">http://test.com?array[]=1&amp;params_in[]=2</a>'
+    )
+
+
 def test_link_query():
     assert (
         linkify("http://xx.com/?test=win")