From 4971a9afe5642569f3dcfcd3972ebb39e88dd457 Mon Sep 17 00:00:00 2001
From: Berker Peksag <berker.peksag@gmail.com>
Date: Tue, 27 Sep 2016 04:59:48 +0300
Subject: [PATCH] Fixed #18119 -- Added a DomainNameValidator validator.

Thanks Claude Paroz for the review.

Co-authored-by: Nina Menezes <77671865+nmenezes0@users.noreply.github.com>
---
 django/core/validators.py | 76 +++++++++++++++++++++++++++++++--------
 docs/ref/validators.txt   | 28 +++++++++++++++
 docs/releases/5.1.txt     |  5 ++-
 tests/validators/tests.py | 56 +++++++++++++++++++++++++++++
 4 files changed, 150 insertions(+), 15 deletions(-)

diff --git a/django/core/validators.py b/django/core/validators.py
index 57940a59da..b1c5c053b8 100644
--- a/django/core/validators.py
+++ b/django/core/validators.py
@@ -65,6 +65,64 @@ class RegexValidator:
         )
 
 
+@deconstructible
+class DomainNameValidator(RegexValidator):
+    message = _("Enter a valid domain name.")
+    ul = "\u00a1-\uffff"  # Unicode letters range (must not be a raw string).
+    # Host patterns.
+    hostname_re = (
+        r"[a-z" + ul + r"0-9](?:[a-z" + ul + r"0-9-]{0,61}[a-z" + ul + r"0-9])?"
+    )
+    # Max length for domain name labels is 63 characters per RFC 1034 sec. 3.1.
+    domain_re = r"(?:\.(?!-)[a-z" + ul + r"0-9-]{1,63}(?<!-))*"
+    # Top-level domain.
+    tld_re = (
+        r"\."  # dot
+        r"(?!-)"  # can't start with a dash
+        r"(?:[a-z" + ul + "-]{2,63}"  # domain label
+        r"|xn--[a-z0-9]{1,59})"  # or punycode label
+        r"(?<!-)"  # can't end with a dash
+        r"\.?"  # may have a trailing dot
+    )
+    ascii_only_hostname_re = r"[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?"
+    ascii_only_domain_re = r"(?:\.(?!-)[a-zA-Z0-9-]{1,63}(?<!-))*"
+    ascii_only_tld_re = (
+        r"\."  # dot
+        r"(?!-)"  # can't start with a dash
+        r"(?:[a-zA-Z0-9-]{2,63})"  # domain label
+        r"(?<!-)"  # can't end with a dash
+        r"\.?"  # may have a trailing dot
+    )
+
+    max_length = 255
+
+    def __init__(self, **kwargs):
+        self.accept_idna = kwargs.pop("accept_idna", True)
+
+        if self.accept_idna:
+            self.regex = _lazy_re_compile(
+                self.hostname_re + self.domain_re + self.tld_re, re.IGNORECASE
+            )
+        else:
+            self.regex = _lazy_re_compile(
+                self.ascii_only_hostname_re
+                + self.ascii_only_domain_re
+                + self.ascii_only_tld_re,
+                re.IGNORECASE,
+            )
+        super().__init__(**kwargs)
+
+    def __call__(self, value):
+        if not isinstance(value, str) or len(value) > self.max_length:
+            raise ValidationError(self.message, code=self.code, params={"value": value})
+        if not self.accept_idna and not value.isascii():
+            raise ValidationError(self.message, code=self.code, params={"value": value})
+        super().__call__(value)
+
+
+validate_domain_name = DomainNameValidator()
+
+
 @deconstructible
 class URLValidator(RegexValidator):
     ul = "\u00a1-\uffff"  # Unicode letters range (must not be a raw string).
@@ -76,20 +134,10 @@ class URLValidator(RegexValidator):
     )
     ipv6_re = r"\[[0-9a-f:.]+\]"  # (simple regex, validated later)
 
-    # Host patterns
-    hostname_re = (
-        r"[a-z" + ul + r"0-9](?:[a-z" + ul + r"0-9-]{0,61}[a-z" + ul + r"0-9])?"
-    )
-    # Max length for domain name labels is 63 characters per RFC 1034 sec. 3.1
-    domain_re = r"(?:\.(?!-)[a-z" + ul + r"0-9-]{1,63}(?<!-))*"
-    tld_re = (
-        r"\."  # dot
-        r"(?!-)"  # can't start with a dash
-        r"(?:[a-z" + ul + "-]{2,63}"  # domain label
-        r"|xn--[a-z0-9]{1,59})"  # or punycode label
-        r"(?<!-)"  # can't end with a dash
-        r"\.?"  # may have a trailing dot
-    )
+    hostname_re = DomainNameValidator.hostname_re
+    domain_re = DomainNameValidator.domain_re
+    tld_re = DomainNameValidator.tld_re
+
     host_re = "(" + hostname_re + domain_re + tld_re + "|localhost)"
 
     regex = _lazy_re_compile(
diff --git a/docs/ref/validators.txt b/docs/ref/validators.txt
index fb69ec8d33..3287d0560e 100644
--- a/docs/ref/validators.txt
+++ b/docs/ref/validators.txt
@@ -159,6 +159,25 @@ to, or in lieu of custom ``field.clean()`` methods.
         validation, so you'd need to add them to the ``allowlist`` as
         necessary.
 
+``DomainNameValidator``
+-----------------------
+
+.. versionadded:: 5.1
+
+.. class:: DomainNameValidator(accept_idna=True, message=None, code=None)
+
+    A :class:`RegexValidator` subclass that ensures a value looks like a domain
+    name. Values longer than 255 characters are always considered invalid. IP
+    addresses are not accepted as valid domain names.
+
+    In addition to the optional arguments of its parent :class:`RegexValidator`
+    class, ``DomainNameValidator`` accepts an extra optional attribute:
+
+    .. attribute:: accept_idna
+
+        Determines whether to accept internationalized domain names, that is,
+        domain names that contain non-ASCII characters. Defaults to ``True``.
+
 ``URLValidator``
 ----------------
 
@@ -201,6 +220,15 @@ to, or in lieu of custom ``field.clean()`` methods.
 
     An :class:`EmailValidator` instance without any customizations.
 
+``validate_domain_name``
+------------------------
+
+.. versionadded:: 5.1
+
+.. data:: validate_domain_name
+
+    A :class:`DomainNameValidator` instance without any customizations.
+
 ``validate_slug``
 -----------------
 
diff --git a/docs/releases/5.1.txt b/docs/releases/5.1.txt
index bb5e4f3fe4..d846788815 100644
--- a/docs/releases/5.1.txt
+++ b/docs/releases/5.1.txt
@@ -363,7 +363,10 @@ Utilities
 Validators
 ~~~~~~~~~~
 
-* ...
+* The new :class:`~django.core.validators.DomainNameValidator` validates domain
+  names, including internationalized domain names. The new
+  :func:`~django.core.validators.validate_domain_name` function returns an
+  instance of :class:`~django.core.validators.DomainNameValidator`.
 
 .. _backwards-incompatible-5.1:
 
diff --git a/tests/validators/tests.py b/tests/validators/tests.py
index 5376517a4a..ba1db5ea46 100644
--- a/tests/validators/tests.py
+++ b/tests/validators/tests.py
@@ -9,6 +9,7 @@ from django.core.files.base import ContentFile
 from django.core.validators import (
     BaseValidator,
     DecimalValidator,
+    DomainNameValidator,
     EmailValidator,
     FileExtensionValidator,
     MaxLengthValidator,
@@ -21,6 +22,7 @@ from django.core.validators import (
     URLValidator,
     int_list_validator,
     validate_comma_separated_integer_list,
+    validate_domain_name,
     validate_email,
     validate_image_file_extension,
     validate_integer,
@@ -618,6 +620,38 @@ TEST_DATA = [
     (ProhibitNullCharactersValidator(), "\x00something", ValidationError),
     (ProhibitNullCharactersValidator(), "something", None),
     (ProhibitNullCharactersValidator(), None, None),
+    (validate_domain_name, "000000.org", None),
+    (validate_domain_name, "python.org", None),
+    (validate_domain_name, "python.co.uk", None),
+    (validate_domain_name, "python.tk", None),
+    (validate_domain_name, "domain.with.idn.tld.उदाहरण.परीक्ष", None),
+    (validate_domain_name, "ıçğü.com", None),
+    (validate_domain_name, "xn--7ca6byfyc.com", None),
+    (validate_domain_name, "hg.python.org", None),
+    (validate_domain_name, "python.xyz", None),
+    (validate_domain_name, "djangoproject.com", None),
+    (validate_domain_name, "DJANGOPROJECT.COM", None),
+    (validate_domain_name, "spam.eggs", None),
+    (validate_domain_name, "python-python.com", None),
+    (validate_domain_name, "python.name.uk", None),
+    (validate_domain_name, "python.tips", None),
+    (validate_domain_name, "http://例子.测试", None),
+    (validate_domain_name, "http://dashinpunytld.xn---c", None),
+    (validate_domain_name, "python..org", ValidationError),
+    (validate_domain_name, "python-.org", ValidationError),
+    (validate_domain_name, "too-long-name." * 20 + "com", ValidationError),
+    (validate_domain_name, "stupid-name试", ValidationError),
+    (validate_domain_name, "255.0.0.0", ValidationError),
+    (validate_domain_name, "fe80::1", ValidationError),
+    (validate_domain_name, "1:2:3:4:5:6:7:8", ValidationError),
+    (DomainNameValidator(accept_idna=False), "non-idna-domain-name-passes.com", None),
+    (
+        DomainNameValidator(accept_idna=False),
+        "domain.with.idn.tld.उदाहरण.परीक्ष",
+        ValidationError,
+    ),
+    (DomainNameValidator(accept_idna=False), "ıçğü.com", ValidationError),
+    (DomainNameValidator(accept_idna=False), "not-domain-name", ValidationError),
 ]
 
 # Add valid and invalid URL tests.
@@ -847,3 +881,25 @@ class TestValidatorEquality(TestCase):
             ProhibitNullCharactersValidator(message="message", code="code1"),
             ProhibitNullCharactersValidator(message="message", code="code2"),
         )
+
+    def test_domain_name_equality(self):
+        self.assertEqual(
+            DomainNameValidator(),
+            DomainNameValidator(),
+        )
+        self.assertNotEqual(
+            DomainNameValidator(),
+            EmailValidator(),
+        )
+        self.assertNotEqual(
+            DomainNameValidator(),
+            DomainNameValidator(code="custom_code"),
+        )
+        self.assertEqual(
+            DomainNameValidator(message="custom error message"),
+            DomainNameValidator(message="custom error message"),
+        )
+        self.assertNotEqual(
+            DomainNameValidator(message="custom error message"),
+            DomainNameValidator(message="custom error message", code="custom_code"),
+        )