Merge pull request #96 from maxmind/ugexe/email-nfc

oschwald · web-flow · commit 28a0fa5bf84d · 2024-04-12T07:38:11.000-07:00
Normalize email addresses to NFC
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -26,6 +26,8 @@
   used. For example, `gmali.com` will become `gmail.com`.
 * When `hash_address` is used, all trailing periods are now removed from an
   email address domain. Previously only a single period was removed.
+* When `hash_address` is used, the local part of an email address is now
+  normalized to NFC.
 
 ## v2.4.0 (2024-01-12)
 
diff --git a/lib/minfraud/components/email.rb b/lib/minfraud/components/email.rb
@@ -88,6 +88,8 @@ def clean_email_address(address)
         local_part, domain = address.split('@', 2)
         return nil if !local_part || !domain
 
+        local_part = local_part.unicode_normalize(:nfc)
+
         domain = clean_domain(domain)
 
         if YAHOO_DOMAINS.key?(domain)
diff --git a/spec/components/email_spec.rb b/spec/components/email_spec.rb
@@ -82,6 +82,8 @@
         { input: 'foo@example.comcom', output: 'foo@example.com' },
         { input: 'foo@example.com.', output: 'foo@example.com' },
         { input: 'foo@example.com...', output: 'foo@example.com' },
+        { input: "example@bu\u0308cher.com", output: 'example@xn--bcher-kva.com' },
+        { input: "example@b\u00FCcher.com", output: 'example@xn--bcher-kva.com' },
       ]
 
       tests.each do |i|
@@ -105,6 +107,14 @@
           input:  'Test+alias@maxmind.com',
           output: '977577b140bfb7c516e4746204fbdb01',
         },
+        {
+          input:  "bu\u0308cher@example.com",
+          output: '53550c712b146287a2d0dd30e5ed6f4b',
+        },
+        {
+          input:  "b\u00FCcher@example.com",
+          output: '53550c712b146287a2d0dd30e5ed6f4b',
+        },
       ]
 
       tests.each do |i|

-Original file line number
+Diff line change
         { input: '[email protected]', output: '[email protected]' },
         { input: '[email protected].', output: '[email protected]' },
         { input: '[email protected]...', output: '[email protected]' },
 +        { input: "example@bu\u0308cher.com", output: '[email protected]' },
 +        { input: "example@b\u00FCcher.com", output: '[email protected]' },
+      ]
       tests.each do |i|
           input:  '[email protected]',
           output: '977577b140bfb7c516e4746204fbdb01',
         },
 +        {
 +          input:  "bu\u0308[email protected]",
 +          output: '53550c712b146287a2d0dd30e5ed6f4b',
 +        },
 +        {
 +          input:  "b\u00FC[email protected]",
 +          output: '53550c712b146287a2d0dd30e5ed6f4b',
 +        },
+      ]
       tests.each do |i|