From 8ff4d48c9ab124356439b7f18ae21f457204736d Mon Sep 17 00:00:00 2001
From: Alex Pott <alex.a.pott@googlemail.com>
Date: Sun, 12 Sep 2021 18:07:29 +0100
Subject: [PATCH] Issue #2016739 by richardbporter, larowlan, aalamaki, afox,
wroxbox, mark_fullmer, mohit_aghera, ayushmishra206, rakesh.gectcr, NikolaAt,
rteijeiro, tanubansal, amietpatial, jibran, alexpott, Wim Leers: Links with
"@" are converted into email addresses even if there is no domain suffix
present
---
core/modules/filter/filter.module | 9 ++++++---
core/modules/filter/tests/filter.url-input.txt | 2 ++
core/modules/filter/tests/filter.url-output.txt | 2 ++
3 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/core/modules/filter/filter.module b/core/modules/filter/filter.module
index 15aad887308d..77e5ff0a189f 100644
--- a/core/modules/filter/filter.module
+++ b/core/modules/filter/filter.module
@@ -490,9 +490,12 @@ function _filter_url($text, $filter) {
// Prepare domain name pattern.
// The ICANN seems to be on track towards accepting more diverse top level
- // domains, so this pattern has been "future-proofed" to allow for TLDs
- // of length 2-64.
+ // domains (TLDs), so this pattern has been "future-proofed" to allow for
+ // TLDs of length 2-64.
$domain = '(?:[\p{L}\p{M}\p{N}._+-]+\.)?[\p{L}\p{M}]{2,64}\b';
+ // Mail domains differ from the generic domain pattern, specifically:
+ // A . character must be present in the string that follows the @ character.
+ $email_domain = '(?:[\p{L}\p{M}\p{N}._+-]+\.)+[\p{L}\p{M}]{2,64}\b';
$ip = '(?:[0-9]{1,3}\.){3}[0-9]{1,3}';
$auth = '[\p{L}\p{M}\p{N}:%_+*~#?&=.,/;-]+@';
$trail = '(' . $valid_url_path . '*)?(\\?' . $valid_url_query_chars . '*' . $valid_url_query_ending_chars . ')?';
@@ -503,7 +506,7 @@ function _filter_url($text, $filter) {
$tasks['_filter_url_parse_full_links'] = $pattern;
// Match email addresses.
- $url_pattern = "[\p{L}\p{M}\p{N}._+-]{1,254}@(?:$domain)";
+ $url_pattern = "[\p{L}\p{M}\p{N}._+-]{1,254}@(?:$email_domain)";
$pattern = "`($url_pattern)`u";
$tasks['_filter_url_parse_email_links'] = $pattern;
diff --git a/core/modules/filter/tests/filter.url-input.txt b/core/modules/filter/tests/filter.url-input.txt
index 92289dcfd169..b41adb3f0392 100644
--- a/core/modules/filter/tests/filter.url-input.txt
+++ b/core/modules/filter/tests/filter.url-input.txt
@@ -9,6 +9,7 @@ This is just a www.test.com. paragraph with person@test.com. some http://www.tes
http://www.test.com
www.test.com
person@test.com
+person@test
<code>www.test.com</code>
http://test.com/?search=test
http://test.com/?search=Test
@@ -28,6 +29,7 @@ The old URL filter has problems with <a title="kind of link www.example.com with
<dt>www.test.com</dt>
<dd>http://www.test.com</dd>
<dd>person@test.com</dd>
+<dd>person@test</dd>
<dt>check www.test.com</dt>
<dd>this with some text around: http://www.test.com not so easy person@test.com now?</dd>
</dl>
diff --git a/core/modules/filter/tests/filter.url-output.txt b/core/modules/filter/tests/filter.url-output.txt
index 814a4ed71766..19acee24a7fb 100644
--- a/core/modules/filter/tests/filter.url-output.txt
+++ b/core/modules/filter/tests/filter.url-output.txt
@@ -9,6 +9,7 @@ This is just a <a href="http://www.test.com">www.test.com</a>. paragraph with <a
<a href="http://www.test.com">http://www.test.com</a>
<a href="http://www.test.com">www.test.com</a>
<a href="mailto:person@test.com">person@test.com</a>
+person@test
<code>www.test.com</code>
<a href="http://test.com/?search=test">http://test.com/?search=test</a>
<a href="http://test.com/?search=Test">http://test.com/?search=Test</a>
@@ -28,6 +29,7 @@ The old URL filter has problems with <a title="kind of link www.example.com with
<dt><a href="http://www.test.com">www.test.com</a></dt>
<dd><a href="http://www.test.com">http://www.test.com</a></dd>
<dd><a href="mailto:person@test.com">person@test.com</a></dd>
+<dd>person@test</dd>
<dt>check <a href="http://www.test.com">www.test.com</a></dt>
<dd>this with some text around: <a href="http://www.test.com">http://www.test.com</a> not so easy <a href="mailto:person@test.com">person@test.com</a> now?</dd>
</dl>
--
GitLab