unicode: Made the serializers unicode-aware. Refs #3878, #4227.

malcolmt · malcolmt · commit 9d001fa7f957 · 2007-05-15T16:14:55.000Z
git-svn-id: http://code.djangoproject.com/svn/django/branches/unicode@5248 bcc190cf-cafb-0310-a4f2-bffc1f526a37
diff --git a/django/core/serializers/__init__.py b/django/core/serializers/__init__.py
@@ -6,7 +6,7 @@
     >>> from django.core import serializers
     >>> json = serializers.serialize("json", some_query_set)
     >>> objects = list(serializers.deserialize("json", json))
-        
+
 To add your own serializers, use the SERIALIZATION_MODULES setting::
 
     SERIALIZATION_MODULES = {
@@ -30,19 +30,19 @@
     import yaml
     BUILTIN_SERIALIZERS["yaml"] = "django.core.serializers.pyyaml"
 except ImportError:
-    pass    
+    pass
 
 _serializers = {}
-        
+
 def register_serializer(format, serializer_module):
     """Register a new serializer by passing in a module name."""
     module = __import__(serializer_module, {}, {}, [''])
     _serializers[format] = module
-    
+
 def unregister_serializer(format):
     """Unregister a given serializer"""
     del _serializers[format]
-    
+
 def get_serializer(format):
     if not _serializers:
         _load_serializers()
@@ -52,12 +52,12 @@ def get_serializer_formats():
     if not _serializers:
         _load_serializers()
     return _serializers.keys()
-    
+
 def get_deserializer(format):
     if not _serializers:
         _load_serializers()
     return _serializers[format].Deserializer
-    
+
 def serialize(format, queryset, **options):
     """
     Serialize a queryset (or any iterator that returns database objects) using
@@ -87,4 +87,4 @@ def _load_serializers():
         register_serializer(format, BUILTIN_SERIALIZERS[format])
     if hasattr(settings, "SERIALIZATION_MODULES"):
         for format in settings.SERIALIZATION_MODULES:
-            register_serializer(format, settings.SERIALIZATION_MODULES[format])
+            register_serializer(format, settings.SERIALIZATION_MODULES[format])
diff --git a/django/core/serializers/base.py b/django/core/serializers/base.py
@@ -7,6 +7,7 @@
 except ImportError:
     from StringIO import StringIO
 from django.db import models
+from django.utils.encoding import smart_str, smart_unicode
 
 class SerializationError(Exception):
     """Something bad happened during serialization."""
@@ -59,7 +60,7 @@ def get_string_value(self, obj, field):
             value = getattr(obj, "get_%s_url" % field.name, lambda: None)()
         else:
             value = field.flatten_data(follow=None, obj=obj).get(field.name, "")
-        return str(value)
+        return smart_unicode(value)
 
     def start_serialization(self):
         """
@@ -154,7 +155,7 @@ def __init__(self, obj, m2m_data=None):
         self.m2m_data = m2m_data
 
     def __repr__(self):
-        return "<DeserializedObject: %s>" % str(self.object)
+        return "<DeserializedObject: %s>" % smart_str(self.object)
 
     def save(self, save_m2m=True):
         self.object.save()
diff --git a/django/core/serializers/python.py b/django/core/serializers/python.py
@@ -7,49 +7,50 @@
 from django.conf import settings
 from django.core.serializers import base
 from django.db import models
+from django.utils.encoding import smart_unicode
 
 class Serializer(base.Serializer):
     """
     Serializes a QuerySet to basic Python objects.
     """
-    
+
     def start_serialization(self):
         self._current = None
         self.objects = []
-        
+
     def end_serialization(self):
         pass
-        
+
     def start_object(self, obj):
         self._current = {}
-        
+
     def end_object(self, obj):
         self.objects.append({
-            "model"  : str(obj._meta),
-            "pk"     : str(obj._get_pk_val()),
+            "model"  : smart_unicode(obj._meta),
+            "pk"     : smart_unicode(obj._get_pk_val()),
             "fields" : self._current
         })
         self._current = None
-        
+
     def handle_field(self, obj, field):
         self._current[field.name] = getattr(obj, field.name)
-        
+
     def handle_fk_field(self, obj, field):
         related = getattr(obj, field.name)
         if related is not None:
             related = getattr(related, field.rel.field_name)
         self._current[field.name] = related
-    
+
     def handle_m2m_field(self, obj, field):
         self._current[field.name] = [related._get_pk_val() for related in getattr(obj, field.name).iterator()]
-    
+
     def getvalue(self):
         return self.objects
 
 def Deserializer(object_list, **options):
     """
     Deserialize simple Python objects back into Django ORM instances.
-    
+
     It's expected that you pass the Python objects themselves (instead of a
     stream or a string) to the constructor
     """
@@ -59,36 +60,30 @@ def Deserializer(object_list, **options):
         Model = _get_model(d["model"])
         data = {Model._meta.pk.attname : Model._meta.pk.to_python(d["pk"])}
         m2m_data = {}
-        
+
         # Handle each field
         for (field_name, field_value) in d["fields"].iteritems():
-            if isinstance(field_value, unicode):
-                field_value = field_value.encode(options.get("encoding", settings.DEFAULT_CHARSET))
-                
+            if isinstance(field_value, str):
+                field_value = smart_unicode(field_value, options.get("encoding", settings.DEFAULT_CHARSET))
+
             field = Model._meta.get_field(field_name)
-            
+
             # Handle M2M relations
             if field.rel and isinstance(field.rel, models.ManyToManyRel):
-                pks = []
                 m2m_convert = field.rel.to._meta.pk.to_python
-                for pk in field_value:
-                    if isinstance(pk, unicode):
-                        pks.append(m2m_convert(pk.encode(options.get("encoding", settings.DEFAULT_CHARSET))))
-                    else:
-                        pks.append(m2m_convert(pk))
-                m2m_data[field.name] = pks
-                
+                m2m_data[field.name] = [m2m_convert(smart_unicode(pk)) for pk in field_value]
+
             # Handle FK fields
             elif field.rel and isinstance(field.rel, models.ManyToOneRel):
                 if field_value:
                     data[field.attname] = field.rel.to._meta.get_field(field.rel.field_name).to_python(field_value)
                 else:
                     data[field.attname] = None
-                    
+
             # Handle all other fields
             else:
                 data[field.name] = field.to_python(field_value)
-                
+
         yield base.DeserializedObject(Model(**data), m2m_data)
 
 def _get_model(model_identifier):
@@ -100,5 +95,5 @@ def _get_model(model_identifier):
     except TypeError:
         Model = None
     if Model is None:
-        raise base.DeserializationError("Invalid model identifier: '%s'" % model_identifier)
+        raise base.DeserializationError(u"Invalid model identifier: '%s'" % model_identifier)
     return Model
diff --git a/django/core/serializers/pyyaml.py b/django/core/serializers/pyyaml.py
@@ -19,7 +19,7 @@ class Serializer(PythonSerializer):
     """
     def end_serialization(self):
         yaml.dump(self.objects, self.stream, **self.options)
-        
+
     def getvalue(self):
         return self.stream.getvalue()
 
@@ -33,4 +33,4 @@ def Deserializer(stream_or_string, **options):
         stream = stream_or_string
     for obj in PythonDeserializer(yaml.load(stream)):
         yield obj
-        
+
diff --git a/django/core/serializers/xml_serializer.py b/django/core/serializers/xml_serializer.py
diff --git a/tests/regressiontests/serializers_regress/tests.py b/tests/regressiontests/serializers_regress/tests.py
