Quote the local part after NFC normalization in validate_email

CHANGELOG.md

@@ -7,6 +7,7 @@ In Development
 * The IPv6 tag of a domain literal is now case-insensitive, as in the RFCs, so `me@[ipv6:::1]` is accepted (and normalized to `me@[IPv6:::1]`) when `allow_domain_literal=True`.
 * Internationalized domains with a right-to-left label are now rejected if any other label breaks the IDNA Bidi Rule (RFC 5893), such as a label starting with a digit. Such domains, like `1.xn--4dbc.com` and `xn--4dbc.1.com` (`1.אב.com` and `אב.1.com` in Unicode), could be displayed identically.
 * Faster validation of domains without Punycode (xn--) labels.
+* With `allow_quoted_local=True`, a quoted local part containing U+037E (Greek question mark), which Unicode NFC normalization changes to a semicolon, was returned without its quotes in the normalized address (e.g. `a;b@example.com`), getting around the re-validation added in 2.2.0. The normalized local part is now quoted according to its normalized form, giving `"a;b"@example.com`.
 
 2.3.0 (August 26, 2025)
 -----------------------

email_validator/validate_email.py

@@ -118,17 +118,22 @@ def validate_email(
     # this library. (UTS #39 seems to require that the *input* be NKFC normalized
     # and has other requirements that are hard to check without additional Unicode
     # data, and I don't know whether the rules really apply in the wild.)
-    normalized_local_part = unicodedata.normalize("NFC", ret.local_part)
-    if normalized_local_part != ret.local_part:
+    #
+    # Normalize the unquoted and unescaped local part, not ret.local_part, and
+    # take the quoting from the second validation. ret.local_part has already
+    # had unnecessary quotes removed, so normalizing it would turn e.g.
+    # "a\u037eb" (quoted, with a Greek question mark) into a;b without quotes.
+    normalized_local_part = unicodedata.normalize("NFC", unescaped_local_part)
+    if normalized_local_part != unescaped_local_part:
         try:
-            validate_email_local_part(normalized_local_part,
-                                      allow_smtputf8=allow_smtputf8,
-                                      allow_empty_local=allow_empty_local,
-                                      quoted_local_part=is_quoted_local_part,
-                                      strict=strict)
+            normalized_local_part_info = validate_email_local_part(normalized_local_part,
+                                                                   allow_smtputf8=allow_smtputf8,
+                                                                   allow_empty_local=allow_empty_local,
+                                                                   quoted_local_part=is_quoted_local_part,
+                                                                   strict=strict)
         except EmailSyntaxError as e:
             raise EmailSyntaxError("After Unicode normalization: " + str(e)) from e
-        ret.local_part = normalized_local_part
+        ret.local_part = normalized_local_part_info["local_part"]
 
     # If a quoted local part isn't allowed but is present, now raise an exception.
     # This is done after any exceptions raised by validate_email_local_part so

tests/test_syntax.py

@@ -389,6 +389,8 @@ def test_email_valid_intl_local_part(email_input: str, output: ValidatedEmail) -
         ('"unnecessarily.quoted.with.unicode.λ"@example.com', 'unnecessarily.quoted.with.unicode.λ'),
         ('"quoted.with..unicode.λ"@example.com', '"quoted.with..unicode.λ"'),
         ('"quoted.with.extraneous.\\escape"@example.com', 'quoted.with.extraneous.escape'),
+        # U+037E (Greek question mark) is valid unquoted but NFC normalizes to a semicolon, which needs the quotes.
+        ('"quoted.after.normalization.\u037e"@example.com', '"quoted.after.normalization.;"'),
     ])
 def test_email_valid_only_if_quoted_local_part(email_input: str, normalized_local_part: str) -> None:
     # These addresses are invalid with the default allow_quoted_local=False option.