Fixed #21242 -- Allowed more IANA schemes in URLValidator

Thanks Sascha Peilicke for the report and initial patch, and
Tim Graham for the review.

django/core/validators.py

@@ -44,7 +44,7 @@ class RegexValidator(object):
 @deconstructible
 class URLValidator(RegexValidator):
     regex = re.compile(
-        r'^(?:http|ftp)s?://'  # http:// or https://
+        r'^(?:[a-z0-9\.\-]*)://'  # scheme is validated separately
         r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}\.?)|'  # domain...
         r'localhost|'  # localhost...
         r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}|'  # ...or ipv4
@@ -52,14 +52,26 @@ class URLValidator(RegexValidator):
         r'(?::\d+)?'  # optional port
         r'(?:/?|[/?]\S+)$', re.IGNORECASE)
     message = _('Enter a valid URL.')
+    schemes = ['http', 'https', 'ftp', 'ftps']
+
+    def __init__(self, schemes=None, **kwargs):
+        super(URLValidator, self).__init__(**kwargs)
+        if schemes is not None:
+            self.schemes = schemes
 
     def __call__(self, value):
+        value = force_text(value)
+        # Check first if the scheme is valid
+        scheme = value.split('://')[0].lower()
+        if scheme not in self.schemes:
+            raise ValidationError(self.message, code=self.code)
+
+        # Then check full URL
         try:
             super(URLValidator, self).__call__(value)
         except ValidationError as e:
             # Trivial case failed. Try for possible IDN domain
             if value:
-                value = force_text(value)
                 scheme, netloc, path, query, fragment = urlsplit(value)
                 try:
                     netloc = netloc.encode('idna').decode('ascii')  # IDN -> ACE

docs/ref/validators.txt

@@ -87,10 +87,24 @@ to, or in lieu of custom ``field.clean()`` methods.
 
 ``URLValidator``
 ----------------
-.. class:: URLValidator()
+.. class:: URLValidator([schemes=None, regex=None, message=None, code=None])
 
     A :class:`RegexValidator` that ensures a value looks like a URL, and raises
-    an error code of ``'invalid'`` if it doesn't.
+    an error code of ``'invalid'`` if it doesn't. In addition to the optional
+    arguments of its parent :class:`RegexValidator` class, ``URLValidator``
+    accepts an extra optional attribute:
+
+    .. attribute:: schemes
+
+        URL/URI scheme list to validate against. If not provided, the default
+        list is ``['http', 'https', 'ftp', 'ftps']``. As a reference, the IANA
+        Web site provides a full list of `valid URI schemes`_.
+
+        .. _valid URI schemes: https://www.iana.org/assignments/uri-schemes/uri-schemes.xhtml
+
+    .. versionchanged:: 1.7
+
+        The optional ``schemes`` attribute was added.
 
 ``validate_email``
 ------------------

docs/releases/1.7.txt

@@ -567,6 +567,13 @@ Tests
 * :meth:`~django.test.TransactionTestCase.assertNumQueries` now prints
   out the list of executed queries if the assertion fails.
 
+Validators
+^^^^^^^^^^
+
+* :class:`~django.core.validators.URLValidator` now accepts an optional
+  ``schemes`` argument which allows customization of the accepted URI schemes
+  (instead of the defaults ``http(s)`` and ``ftp(s)``).
+
 Backwards incompatible changes in 1.7
 =====================================
 

tests/validators/tests.py

@@ -18,6 +18,7 @@ from django.test.utils import str_prefix
 
 
 NOW = datetime.now()
+EXTENDED_SCHEMES = ['http', 'https', 'ftp', 'ftps', 'git', 'file']
 
 TEST_DATA = (
     # (validator, value, expected),
@@ -141,6 +142,7 @@ TEST_DATA = (
     (MinLengthValidator(10), '', ValidationError),
 
     (URLValidator(), 'http://www.djangoproject.com/', None),
+    (URLValidator(), 'HTTP://WWW.DJANGOPROJECT.COM/', None),
     (URLValidator(), 'http://localhost/', None),
     (URLValidator(), 'http://example.com/', None),
     (URLValidator(), 'http://www.example.com/', None),
@@ -155,6 +157,8 @@ TEST_DATA = (
     (URLValidator(), 'https://example.com/', None),
     (URLValidator(), 'ftp://example.com/', None),
     (URLValidator(), 'ftps://example.com/', None),
+    (URLValidator(EXTENDED_SCHEMES), 'file://localhost/path', None),
+    (URLValidator(EXTENDED_SCHEMES), 'git://example.com/', None),
 
     (URLValidator(), 'foo', ValidationError),
     (URLValidator(), 'http://', ValidationError),
@@ -165,6 +169,9 @@ TEST_DATA = (
     (URLValidator(), 'http://-invalid.com', ValidationError),
     (URLValidator(), 'http://inv-.alid-.com', ValidationError),
     (URLValidator(), 'http://inv-.-alid.com', ValidationError),
+    (URLValidator(), 'file://localhost/path', ValidationError),
+    (URLValidator(), 'git://example.com/', ValidationError),
+    (URLValidator(EXTENDED_SCHEMES), 'git://-invalid.com', ValidationError),
 
     (BaseValidator(True), True, None),
     (BaseValidator(True), False, ValidationError),