Commit 7f5d9ad6 authored by Jannis Leidel's avatar Jannis Leidel
Browse files

Fixed #9764 - Updated EmailField and URLField to support IDN... 

Fixed #9764 - Updated EmailField and URLField to support IDN (Internationalized Domain Names). Thanks, UloPe.

git-svn-id: http://code.djangoproject.com/svn/django/trunk@12474 bcc190cf-cafb-0310-a4f2-bffc1f526a37
parent 0d4726f5

django/core/validators.py

+42 −2
Original line number Diff line number Diff line 
import re

import urlparse



from django.core.exceptions import ValidationError

from django.utils.translation import ugettext_lazy as _
@@ -52,7 +53,29 @@ class URLValidator(RegexValidator): 
        self.user_agent = validator_user_agent



    def __call__(self, value):

        try:

            super(URLValidator, self).__call__(value)

        except ValidationError, e:

            # Trivial case failed. Try for possible IDN domain

            if value:

                original = value

                value = smart_unicode(value)

                splitted = urlparse.urlsplit(value)

                try:

                    netloc_ace = splitted[1].encode('idna') # IDN -> ACE

                except UnicodeError: # invalid domain part

                    raise e

                value = value.replace(splitted[1], netloc_ace)

                # If no URL path given, assume /

                if not splitted[2]:

                    value += u'/'

                super(URLValidator, self).__call__(value)

                # After validation revert ACE encoded domain-part to

                # original (IDN) value as suggested by RFC 3490

                value = original

            else:

                raise



        if self.verify_exists:

            import urllib2

            headers = {
@@ -77,12 +100,29 @@ def validate_integer(value): 
    except (ValueError, TypeError), e:

        raise ValidationError('')



class EmailValidator(RegexValidator):



    def __call__(self, value):

        try:

            super(EmailValidator, self).__call__(value)

        except ValidationError, e:

            # Trivial case failed. Try for possible IDN domain-part

            if value and u'@' in value:

                parts = value.split(u'@')

                domain_part = parts[-1]

                try:

                    parts[-1] = parts[-1].encode('idna')

                except UnicodeError:

                    raise e

                super(EmailValidator, self).__call__(u'@'.join(parts))

            else:

                raise



email_re = re.compile(

    r"(^[-!#$%&'*+/=?^_`{}|~0-9A-Z]+(\.[-!#$%&'*+/=?^_`{}|~0-9A-Z]+)*"  # dot-atom

    r'|^"([\001-\010\013\014\016-\037!#-\[\]-\177]|\\[\001-011\013\014\016-\177])*"' # quoted-string

    r')@(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+[A-Z]{2,6}\.?$', re.IGNORECASE)  # domain

validate_email = RegexValidator(email_re, _(u'Enter a valid e-mail address.'), 'invalid')

validate_email = EmailValidator(email_re, _(u'Enter a valid e-mail address.'), 'invalid')



slug_re = re.compile(r'^[-\w]+$')

validate_slug = RegexValidator(slug_re, _(u"Enter a valid 'slug' consisting of letters, numbers, underscores or hyphens."), 'invalid')

docs/ref/forms/fields.txt

+11 −0
Original line number Diff line number Diff line
@@ -482,6 +482,11 @@ Has two optional arguments for validation, ``max_length`` and ``min_length``. 
If provided, these arguments ensure that the string is at most or at least the

given length.



.. versionchanged:: 1.2

   The EmailField previously did not recognize e-mail addresses as valid that

   contained an IDN (Internationalized Domain Name; a domain containing

   unicode characters) domain part. This has now been corrected.



``FileField``

~~~~~~~~~~~~~
@@ -707,6 +712,12 @@ Takes the following optional arguments: 
    String used as the user-agent used when checking for a URL's existence.

    Defaults to the value of the ``URL_VALIDATOR_USER_AGENT`` setting.



.. versionchanged:: 1.2

   The URLField previously did not recognize URLs as valid that contained an IDN

   (Internationalized Domain Name; a domain name containing unicode characters)

   domain name. This has now been corrected.





Slightly complex built-in ``Field`` classes

-------------------------------------------

tests/regressiontests/forms/fields.py

+3 −0
Original line number Diff line number Diff line
@@ -408,6 +408,8 @@ class FieldsTests(TestCase): 
        self.assertRaisesErrorWithMessage(ValidationError, "[u'Enter a valid e-mail address.']", f.clean, 'example@inv-.-alid.com')

        self.assertEqual(u'example@valid-----hyphens.com', f.clean('example@valid-----hyphens.com'))

        self.assertEqual(u'example@valid-with-hyphens.com', f.clean('example@valid-with-hyphens.com'))

        self.assertRaisesErrorWithMessage(ValidationError, "[u'Enter a valid e-mail address.']", f.clean, 'example@.com')

        self.assertEqual(u'local@domain.with.idn.xyz\xe4\xf6\xfc\xdfabc.part.com', f.clean('local@domain.with.idn.xyzäöüßabc.part.com'))



    def test_email_regexp_for_performance(self):

        f = EmailField()
@@ -489,6 +491,7 @@ class FieldsTests(TestCase): 
        self.assertRaisesErrorWithMessage(ValidationError, "[u'Enter a valid URL.']", f.clean, 'http://inv-.alid-.com')

        self.assertRaisesErrorWithMessage(ValidationError, "[u'Enter a valid URL.']", f.clean, 'http://inv-.-alid.com')

        self.assertEqual(u'http://valid-----hyphens.com/', f.clean('http://valid-----hyphens.com'))

        self.assertEqual(u'http://some.idn.xyz\xe4\xf6\xfc\xdfabc.domain.com:123/blah', f.clean('http://some.idn.xyzäöüßabc.domain.com:123/blah'))



    def test_url_regex_ticket11198(self):

        f = URLField()