Use unicode internally everywhere for 'repr'

8 files changed, 67 insertions, 21 deletions

diff --git a/rest_framework/compat.py b/rest_framework/compat.py
index 71520b92..69fdd793 100644
--- a/rest_framework/compat.py
+++ b/rest_framework/compat.py
@@ -16,6 +16,23 @@ from django.utils import six
 import django
 
 
+def unicode_repr(instance):
+    # Get the repr of an instance, but ensure it is a unicode string
+    # on both python 3 (already the case) and 2 (not the case).
+    if six.PY2:
+        repr(instance).decode('utf-8')
+    return repr(instance)
+
+
+def unicode_to_repr(value):
+    # Coerce a unicode string to the correct repr return type, depending on
+    # the Python version. We wrap all our `__repr__` implementations with
+    # this and then use unicode throughout internally.
+    if six.PY2:
+        return value.encode('utf-8')
+    return value
+
+
 # OrderedDict only available in Python 2.7.
 # This will always be the case in Django 1.7 and above, as these versions
 # no longer support Python 2.6.
diff --git a/rest_framework/fields.py b/rest_framework/fields.py
index 205efd2f..1a4712d8 100644
--- a/rest_framework/fields.py
+++ b/rest_framework/fields.py
@@ -1,3 +1,4 @@
+from __future__ import unicode_literals
 from django.conf import settings
 from django.core.exceptions import ObjectDoesNotExist
 from django.core.exceptions import ValidationError as DjangoValidationError
@@ -10,7 +11,8 @@ from django.utils.translation import ugettext_lazy as _
 from rest_framework import ISO_8601
 from rest_framework.compat import (
     EmailValidator, MinValueValidator, MaxValueValidator,
-    MinLengthValidator, MaxLengthValidator, URLValidator, OrderedDict
+    MinLengthValidator, MaxLengthValidator, URLValidator, OrderedDict,
+    unicode_repr, unicode_to_repr
 )
 from rest_framework.exceptions import ValidationError
 from rest_framework.settings import api_settings
@@ -113,7 +115,9 @@ class CreateOnlyDefault:
         return self.default
 
     def __repr__(self):
-        return '%s(%s)' % (self.__class__.__name__, repr(self.default))
+        return unicode_to_repr(
+            '%s(%s)' % (self.__class__.__name__, unicode_repr(self.default))
+        )
 
 
 class CurrentUserDefault:
@@ -124,7 +128,7 @@ class CurrentUserDefault:
         return self.user
 
     def __repr__(self):
-        return '%s()' % self.__class__.__name__
+        return unicode_to_repr('%s()' % self.__class__.__name__)
 
 
 class SkipField(Exception):
@@ -453,7 +457,7 @@ class Field(object):
         This allows us to create descriptive representations for serializer
         instances that show all the declared fields on the serializer.
         """
-        return representation.field_repr(self)
+        return unicode_to_repr(representation.field_repr(self))
 
 
 # Boolean types...
diff --git a/rest_framework/serializers.py b/rest_framework/serializers.py
index 5adbca3b..e9860a2f 100644
--- a/rest_framework/serializers.py
+++ b/rest_framework/serializers.py
@@ -10,12 +10,11 @@ python primitives.
 2. The process of marshalling between python primitives and request and
 response content is handled by parsers and renderers.
 """
-import warnings
-
+from __future__ import unicode_literals
 from django.db import models
 from django.db.models.fields import FieldDoesNotExist
 from django.utils.translation import ugettext_lazy as _
-
+from rest_framework.compat import unicode_to_repr
 from rest_framework.utils import model_meta
 from rest_framework.utils.field_mapping import (
     get_url_kwargs, get_field_kwargs,
@@ -29,6 +28,7 @@ from rest_framework.validators import (
     UniqueForDateValidator, UniqueForMonthValidator, UniqueForYearValidator,
     UniqueTogetherValidator
 )
+import warnings
 
 
 # Note: We do the following so that users of the framework can use this style:
@@ -396,7 +396,7 @@ class Serializer(BaseSerializer):
         return attrs
 
     def __repr__(self):
-        return representation.serializer_repr(self, indent=1)
+        return unicode_to_repr(representation.serializer_repr(self, indent=1))
 
     # The following are used for accessing `BoundField` instances on the
     # serializer, for the purposes of presenting a form-like API onto the
@@ -564,7 +564,7 @@ class ListSerializer(BaseSerializer):
         return self.instance
 
     def __repr__(self):
-        return representation.list_repr(self, indent=1)
+        return unicode_to_repr(representation.list_repr(self, indent=1))
 
     # Include a backlink to the serializer class on return objects.
     # Allows renderers such as HTMLFormRenderer to get the full field info.
diff --git a/rest_framework/utils/representation.py b/rest_framework/utils/representation.py
index 3f17a8b9..0d829a83 100644
--- a/rest_framework/utils/representation.py
+++ b/rest_framework/utils/representation.py
@@ -2,9 +2,11 @@
 Helper functions for creating user-friendly representations
 of serializer classes and serializer fields.
 """
+from __future__ import unicode_literals
 from django.db import models
 from django.utils.encoding import force_text
 from django.utils.functional import Promise
+from rest_framework.compat import unicode_repr
 import re
 
 
@@ -24,10 +26,11 @@ def smart_repr(value):
     if isinstance(value, Promise) and value._delegate_text:
         value = force_text(value)
 
-    value = repr(value)
+    value = unicode_repr(value)
 
     # Representations like u'help text'
     # should simply be presented as 'help text'
+    print type(value), value
     if value.startswith("u'") and value.endswith("'"):
         return value[1:]
 
diff --git a/rest_framework/utils/serializer_helpers.py b/rest_framework/utils/serializer_helpers.py
index 277cf649..65a04d06 100644
--- a/rest_framework/utils/serializer_helpers.py
+++ b/rest_framework/utils/serializer_helpers.py
@@ -1,5 +1,6 @@
+from __future__ import unicode_literals
 import collections
-from rest_framework.compat import OrderedDict
+from rest_framework.compat import OrderedDict, unicode_to_repr
 
 
 class ReturnDict(OrderedDict):
@@ -47,9 +48,9 @@ class BoundField(object):
         return self._field.__class__
 
     def __repr__(self):
-        return '<%s value=%s errors=%s>' % (
+        return unicode_to_repr('<%s value=%s errors=%s>' % (
             self.__class__.__name__, self.value, self.errors
-        )
+        ))
 
 
 class NestedBoundField(BoundField):
diff --git a/rest_framework/validators.py b/rest_framework/validators.py
index 63eb7b22..e3719b8d 100644
--- a/rest_framework/validators.py
+++ b/rest_framework/validators.py
@@ -6,7 +6,9 @@ This gives us better separation of concerns, allows us to use single-step
 object creation, and makes it possible to switch between using the implicit
 `ModelSerializer` class and an equivalent explicit `Serializer` class.
 """
+from __future__ import unicode_literals
 from django.utils.translation import ugettext_lazy as _
+from rest_framework.compat import unicode_to_repr
 from rest_framework.exceptions import ValidationError
 from rest_framework.utils.representation import smart_repr
 
@@ -59,10 +61,10 @@ class UniqueValidator:
             raise ValidationError(self.message)
 
     def __repr__(self):
-        return '<%s(queryset=%s)>' % (
+        return unicode_to_repr('<%s(queryset=%s)>' % (
             self.__class__.__name__,
             smart_repr(self.queryset)
-        )
+        ))
 
 
 class UniqueTogetherValidator:
@@ -141,11 +143,11 @@ class UniqueTogetherValidator:
             raise ValidationError(self.message.format(field_names=field_names))
 
     def __repr__(self):
-        return '<%s(queryset=%s, fields=%s)>' % (
+        return unicode_to_repr('<%s(queryset=%s, fields=%s)>' % (
             self.__class__.__name__,
             smart_repr(self.queryset),
             smart_repr(self.fields)
-        )
+        ))
 
 
 class BaseUniqueForValidator:
@@ -205,12 +207,12 @@ class BaseUniqueForValidator:
             raise ValidationError({self.field: message})
 
     def __repr__(self):
-        return '<%s(queryset=%s, field=%s, date_field=%s)>' % (
+        return unicode_to_repr('<%s(queryset=%s, field=%s, date_field=%s)>' % (
             self.__class__.__name__,
             smart_repr(self.queryset),
             smart_repr(self.field),
             smart_repr(self.date_field)
-        )
+        ))
 
 
 class UniqueForDateValidator(BaseUniqueForValidator):
diff --git a/tests/test_fields.py b/tests/test_fields.py
index 3f4e65f2..c20bdd8c 100644
--- a/tests/test_fields.py
+++ b/tests/test_fields.py
@@ -62,7 +62,7 @@ class TestEmpty:
         """
         field = serializers.CharField(allow_blank=True)
         output = field.run_validation('')
-        assert output is ''
+        assert output == ''
 
     def test_default(self):
         """
@@ -817,7 +817,7 @@ class TestChoiceField(FieldValues):
             ]
         )
         output = field.run_validation('')
-        assert output is ''
+        assert output == ''
 
 
 class TestChoiceFieldWithType(FieldValues):
diff --git a/tests/test_serializer.py b/tests/test_serializer.py
index 56b39095..c17b6d8c 100644
--- a/tests/test_serializer.py
+++ b/tests/test_serializer.py
@@ -1,5 +1,7 @@
+# coding: utf-8
 from __future__ import unicode_literals
 from rest_framework import serializers
+from rest_framework.compat import unicode_repr
 import pytest
 
 
@@ -197,3 +199,20 @@ class TestIncorrectlyConfigured:
             "The serializer field might be named incorrectly and not match any attribute or key on the `ExampleObject` instance.\n"
             "Original exception text was:"
         )
+
+
+class TestUnicodeRepr:
+    def test_unicode_repr(self):
+        class ExampleSerializer(serializers.Serializer):
+            example = serializers.CharField()
+
+        class ExampleObject:
+            def __init__(self):
+                self.example = '한국'
+
+            def __repr__(self):
+                return unicode_repr(self.example)
+
+        instance = ExampleObject()
+        serializer = ExampleSerializer(instance)
+        repr(serializer)  # Should not error.