From b2dc66448503c2120d943a2f282eab235afc67ba Mon Sep 17 00:00:00 2001 From: Tom Christie Date: Tue, 19 Mar 2013 14:26:48 +0000 Subject: [PATCH 1/4] Basic bulk create and bulk update --- docs/api-guide/serializers.md | 17 +- rest_framework/serializers.py | 38 ++++ rest_framework/tests/serializer.py | 2 +- .../tests/serializer_bulk_update.py | 174 ++++++++++++++++++ 4 files changed, 223 insertions(+), 8 deletions(-) create mode 100644 rest_framework/tests/serializer_bulk_update.py diff --git a/docs/api-guide/serializers.md b/docs/api-guide/serializers.md index 42edf9af1..ee7208a36 100644 --- a/docs/api-guide/serializers.md +++ b/docs/api-guide/serializers.md @@ -37,9 +37,6 @@ Declaring a serializer looks very similar to declaring a form: """ Given a dictionary of deserialized field values, either update an existing model instance, or create a new model instance. - - Note that if we don't define this method, then deserializing - data will simply return a dictionary of items. """ if instance is not None: instance.title = attrs.get('title', instance.title) @@ -48,7 +45,9 @@ Declaring a serializer looks very similar to declaring a form: return instance return Comment(**attrs) -The first part of serializer class defines the fields that get serialized/deserialized. The `restore_object` method defines how fully fledged instances get created when deserializing data. The `restore_object` method is optional, and is only required if we want our serializer to support deserialization. +The first part of serializer class defines the fields that get serialized/deserialized. The `restore_object` method defines how fully fledged instances get created when deserializing data. + +The `restore_object` method is optional, and is only required if we want our serializer to support deserialization into fully fledged object instances. If we don't define this method, then deserializing data will simply return a dictionary of items. ## Serializing objects @@ -88,18 +87,22 @@ By default, serializers must be passed values for all required fields or they wi serializer = CommentSerializer(comment, data={'content': u'foo bar'}, partial=True) # Update `instance` with partial data -## Serializing querysets +## Serializing multiple objects -To serialize a queryset instead of an object instance, you should pass the `many=True` flag when instantiating the serializer. +To serialize a queryset or list of objects instead of a single object instance, you should pass the `many=True` flag when instantiating the serializer. queryset = Comment.objects.all() serializer = CommentSerializer(queryset, many=True) serializer.data - # [{'email': u'leila@example.com', 'content': u'foo bar', 'created': datetime.datetime(2012, 8, 22, 16, 20, 9, 822774)}, {'email': u'jamie@example.com', 'content': u'baz', 'created': datetime.datetime(2013, 1, 12, 16, 12, 45, 104445)}] + # [ + # {'email': u'leila@example.com', 'content': u'foo bar', 'created': datetime.datetime(2012, 8, 22, 16, 20, 9, 822774)}, + # {'email': u'jamie@example.com', 'content': u'baz', 'created': datetime.datetime(2013, 1, 12, 16, 12, 45, 104445)} + # ] ## Validation When deserializing data, you always need to call `is_valid()` before attempting to access the deserialized object. If any validation errors occur, the `.errors` property will contain a dictionary representing the resulting error messages. + Each key in the dictionary will be the field name, and the values will be lists of strings of any error messages corresponding to that field. The `non_field_errors` key may also be present, and will list any general validation errors. When deserializing a list of items, errors will be returned as a list of dictionaries representing each of the deserialized items. diff --git a/rest_framework/serializers.py b/rest_framework/serializers.py index 4fe857a61..34120dc68 100644 --- a/rest_framework/serializers.py +++ b/rest_framework/serializers.py @@ -128,6 +128,7 @@ class BaseSerializer(Field): self._data = None self._files = None self._errors = None + self._deleted = None ##### # Methods to determine which fields to use when (de)serializing objects. @@ -331,6 +332,13 @@ class BaseSerializer(Field): return [self.to_native(item) for item in obj] return self.to_native(obj) + def get_identity(self, data): + """ + This hook is required for bulk update. + It is used to determine the canonical identity of a given object. + """ + return data.get('id') + @property def errors(self): """ @@ -352,9 +360,32 @@ class BaseSerializer(Field): if many: ret = [] errors = [] + update = self.object is not None + + if update: + # If this is a bulk update we need to map all the objects + # to a canonical identity so we can determine which + # individual object is being updated for each item in the + # incoming data + objects = self.object + identities = [self.get_identity(self.to_native(obj)) for obj in objects] + identity_to_objects = dict(zip(identities, objects)) + for item in data: + if update: + # Determine which object we're updating + try: + identity = self.get_identity(item) + except: + self.object = None + else: + self.object = identity_to_objects.pop(identity, None) + ret.append(self.from_native(item, None)) errors.append(self._errors) + + if update: + self._deleted = identity_to_objects.values() self._errors = any(errors) and errors or [] else: ret = self.from_native(data, files) @@ -394,6 +425,9 @@ class BaseSerializer(Field): def save_object(self, obj, **kwargs): obj.save(**kwargs) + def delete_object(self, obj): + obj.delete() + def save(self, **kwargs): """ Save the deserialized object and return it. @@ -402,6 +436,10 @@ class BaseSerializer(Field): [self.save_object(item, **kwargs) for item in self.object] else: self.save_object(self.object, **kwargs) + + if self._deleted: + [self.delete_object(item) for item in self._deleted] + return self.object diff --git a/rest_framework/tests/serializer.py b/rest_framework/tests/serializer.py index beb372c2b..9c0fdd787 100644 --- a/rest_framework/tests/serializer.py +++ b/rest_framework/tests/serializer.py @@ -266,7 +266,7 @@ class ValidationTests(TestCase): Data of the wrong type is not valid. """ data = ['i am', 'a', 'list'] - serializer = CommentSerializer(self.comment, data=data, many=True) + serializer = CommentSerializer([self.comment], data=data, many=True) self.assertEqual(serializer.is_valid(), False) self.assertTrue(isinstance(serializer.errors, list)) diff --git a/rest_framework/tests/serializer_bulk_update.py b/rest_framework/tests/serializer_bulk_update.py new file mode 100644 index 000000000..66fca8835 --- /dev/null +++ b/rest_framework/tests/serializer_bulk_update.py @@ -0,0 +1,174 @@ +""" +Tests to cover bulk create and update using serializers. +""" +from __future__ import unicode_literals +from django.test import TestCase +from rest_framework import serializers + + +class BulkCreateSerializerTests(TestCase): + + def setUp(self): + class BookSerializer(serializers.Serializer): + id = serializers.IntegerField() + title = serializers.CharField(max_length=100) + author = serializers.CharField(max_length=100) + + self.BookSerializer = BookSerializer + + def test_bulk_create_success(self): + """ + Correct bulk update serialization should return the input data. + """ + + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 1, + 'title': 'If this is a man', + 'author': 'Primo Levi' + }, { + 'id': 2, + 'title': 'The wind-up bird chronicle', + 'author': 'Haruki Murakami' + } + ] + + serializer = self.BookSerializer(data=data, many=True) + self.assertEqual(serializer.is_valid(), True) + self.assertEqual(serializer.object, data) + + def test_bulk_create_errors(self): + """ + Correct bulk update serialization should return the input data. + """ + + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 1, + 'title': 'If this is a man', + 'author': 'Primo Levi' + }, { + 'id': 'foo', + 'title': 'The wind-up bird chronicle', + 'author': 'Haruki Murakami' + } + ] + expected_errors = [ + {}, + {}, + {'id': ['Enter a whole number.']} + ] + + serializer = self.BookSerializer(data=data, many=True) + self.assertEqual(serializer.is_valid(), False) + self.assertEqual(serializer.errors, expected_errors) + + +class BulkUpdateSerializerTests(TestCase): + + def setUp(self): + class Book(object): + object_map = {} + + def __init__(self, id, title, author): + self.id = id + self.title = title + self.author = author + + def save(self): + Book.object_map[self.id] = self + + def delete(self): + del Book.object_map[self.id] + + class BookSerializer(serializers.Serializer): + id = serializers.IntegerField() + title = serializers.CharField(max_length=100) + author = serializers.CharField(max_length=100) + + def restore_object(self, attrs, instance=None): + if instance: + instance.id = attrs['id'] + instance.title = attrs['title'] + instance.author = attrs['author'] + return instance + return Book(**attrs) + + self.Book = Book + self.BookSerializer = BookSerializer + + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 1, + 'title': 'If this is a man', + 'author': 'Primo Levi' + }, { + 'id': 2, + 'title': 'The wind-up bird chronicle', + 'author': 'Haruki Murakami' + } + ] + + for item in data: + book = Book(item['id'], item['title'], item['author']) + book.save() + + def books(self): + return self.Book.object_map.values() + + def test_bulk_update_success(self): + """ + Correct bulk update serialization should return the input data. + """ + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 2, + 'title': 'Kafka on the shore', + 'author': 'Haruki Murakami' + } + ] + serializer = self.BookSerializer(self.books(), data=data, many=True) + self.assertEqual(serializer.is_valid(), True) + self.assertEqual(serializer.data, data) + serializer.save() + new_data = self.BookSerializer(self.books(), many=True).data + self.assertEqual(data, new_data) + + def test_bulk_update_error(self): + """ + Correct bulk update serialization should return the input data. + """ + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 'foo', + 'title': 'Kafka on the shore', + 'author': 'Haruki Murakami' + } + ] + expected_errors = [ + {}, + {'id': ['Enter a whole number.']} + ] + serializer = self.BookSerializer(self.books(), data=data, many=True) + self.assertEqual(serializer.is_valid(), False) + self.assertEqual(serializer.errors, expected_errors) From c32d9969acaebc83ad1e2dd888d4a2829c35571e Mon Sep 17 00:00:00 2001 From: Tom Christie Date: Fri, 22 Mar 2013 12:33:09 +0000 Subject: [PATCH 2/4] Add extra tests for errors from incorrect data with multiple create/update --- rest_framework/serializers.py | 36 ++++++----- .../tests/serializer_bulk_update.py | 60 ++++++++++++++++++- 2 files changed, 81 insertions(+), 15 deletions(-) diff --git a/rest_framework/serializers.py b/rest_framework/serializers.py index 34120dc68..3029cf1c1 100644 --- a/rest_framework/serializers.py +++ b/rest_framework/serializers.py @@ -371,22 +371,30 @@ class BaseSerializer(Field): identities = [self.get_identity(self.to_native(obj)) for obj in objects] identity_to_objects = dict(zip(identities, objects)) - for item in data: + try: + iter(data) + if isinstance(data, dict): + raise TypeError + except TypeError: + self._errors = {'non_field_errors': ['Expected a list of items']} + else: + for item in data: + if update: + # Determine which object we're updating + try: + identity = self.get_identity(item) + except: + self.object = None + else: + self.object = identity_to_objects.pop(identity, None) + + ret.append(self.from_native(item, None)) + errors.append(self._errors) + if update: - # Determine which object we're updating - try: - identity = self.get_identity(item) - except: - self.object = None - else: - self.object = identity_to_objects.pop(identity, None) + self._deleted = identity_to_objects.values() - ret.append(self.from_native(item, None)) - errors.append(self._errors) - - if update: - self._deleted = identity_to_objects.values() - self._errors = any(errors) and errors or [] + self._errors = any(errors) and errors or [] else: ret = self.from_native(data, files) diff --git a/rest_framework/tests/serializer_bulk_update.py b/rest_framework/tests/serializer_bulk_update.py index 66fca8835..2f4164880 100644 --- a/rest_framework/tests/serializer_bulk_update.py +++ b/rest_framework/tests/serializer_bulk_update.py @@ -7,6 +7,9 @@ from rest_framework import serializers class BulkCreateSerializerTests(TestCase): + """ + Creating multiple instances using serializers. + """ def setUp(self): class BookSerializer(serializers.Serializer): @@ -71,11 +74,63 @@ class BulkCreateSerializerTests(TestCase): self.assertEqual(serializer.is_valid(), False) self.assertEqual(serializer.errors, expected_errors) + def test_invalid_list_datatype(self): + """ + Data containing list of incorrect data type should return errors. + """ + data = ['foo', 'bar', 'baz'] + serializer = self.BookSerializer(data=data, many=True) + self.assertEqual(serializer.is_valid(), False) + + expected_errors = [ + {'non_field_errors': ['Invalid data']}, + {'non_field_errors': ['Invalid data']}, + {'non_field_errors': ['Invalid data']} + ] + + self.assertEqual(serializer.errors, expected_errors) + + def test_invalid_single_datatype(self): + """ + Data containing a single incorrect data type should return errors. + """ + data = 123 + serializer = self.BookSerializer(data=data, many=True) + self.assertEqual(serializer.is_valid(), False) + + expected_errors = {'non_field_errors': ['Expected a list of items']} + + self.assertEqual(serializer.errors, expected_errors) + + def test_invalid_single_object(self): + """ + Data containing only a single object, instead of a list of objects + should return errors. + """ + data = { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + } + serializer = self.BookSerializer(data=data, many=True) + self.assertEqual(serializer.is_valid(), False) + + expected_errors = {'non_field_errors': ['Expected a list of items']} + + self.assertEqual(serializer.errors, expected_errors) + class BulkUpdateSerializerTests(TestCase): + """ + Updating multiple instances using serializers. + """ def setUp(self): class Book(object): + """ + A data type that can be persisted to a mock storage backend + with `.save()` and `.delete()`. + """ object_map = {} def __init__(self, id, title, author): @@ -126,6 +181,9 @@ class BulkUpdateSerializerTests(TestCase): book.save() def books(self): + """ + Return all the objects in the mock storage backend. + """ return self.Book.object_map.values() def test_bulk_update_success(self): @@ -152,7 +210,7 @@ class BulkUpdateSerializerTests(TestCase): def test_bulk_update_error(self): """ - Correct bulk update serialization should return the input data. + Incorrect bulk update serialization should return error data. """ data = [ { From b4210f9a56fb7f8913f674aecb304da01e2e9f64 Mon Sep 17 00:00:00 2001 From: Tom Christie Date: Fri, 22 Mar 2013 12:46:03 +0000 Subject: [PATCH 3/4] Test moved out of serializer.py into serializer_bulk_update --- rest_framework/tests/serializer.py | 28 ---------------------------- 1 file changed, 28 deletions(-) diff --git a/rest_framework/tests/serializer.py b/rest_framework/tests/serializer.py index 9c0fdd787..1ee671dcb 100644 --- a/rest_framework/tests/serializer.py +++ b/rest_framework/tests/serializer.py @@ -261,34 +261,6 @@ class ValidationTests(TestCase): self.assertEqual(serializer.is_valid(), True) self.assertEqual(serializer.errors, {}) - def test_bad_type_data_is_false(self): - """ - Data of the wrong type is not valid. - """ - data = ['i am', 'a', 'list'] - serializer = CommentSerializer([self.comment], data=data, many=True) - self.assertEqual(serializer.is_valid(), False) - self.assertTrue(isinstance(serializer.errors, list)) - - self.assertEqual( - serializer.errors, - [ - {'non_field_errors': ['Invalid data']}, - {'non_field_errors': ['Invalid data']}, - {'non_field_errors': ['Invalid data']} - ] - ) - - data = 'and i am a string' - serializer = CommentSerializer(self.comment, data=data) - self.assertEqual(serializer.is_valid(), False) - self.assertEqual(serializer.errors, {'non_field_errors': ['Invalid data']}) - - data = 42 - serializer = CommentSerializer(self.comment, data=data) - self.assertEqual(serializer.is_valid(), False) - self.assertEqual(serializer.errors, {'non_field_errors': ['Invalid data']}) - def test_cross_field_validation(self): class CommentSerializerWithCrossFieldValidator(CommentSerializer): From 13794baf7016f7d44daffb55d29e3dbc56f7612d Mon Sep 17 00:00:00 2001 From: Tom Christie Date: Fri, 22 Mar 2013 17:01:06 +0000 Subject: [PATCH 4/4] Bit of extra tidying and plenty of docs --- docs/api-guide/fields.md | 43 ++++ docs/api-guide/serializers.md | 189 ++++++++++++------ rest_framework/serializers.py | 49 +++-- .../tests/serializer_bulk_update.py | 26 ++- 4 files changed, 225 insertions(+), 82 deletions(-) diff --git a/docs/api-guide/fields.md b/docs/api-guide/fields.md index 9a745cf19..4d73eec76 100644 --- a/docs/api-guide/fields.md +++ b/docs/api-guide/fields.md @@ -273,6 +273,49 @@ Django's regular [FILE_UPLOAD_HANDLERS] are used for handling uploaded files. --- +# Custom fields + +If you want to create a custom field, you'll probably want to override either one or both of the `.to_native()` and `.from_native()` methods. These two methods are used to convert between the intial datatype, and a primative, serializable datatype. Primative datatypes may be any of a number, string, date/time/datetime or None. They may also be any list or dictionary like object that only contains other primative objects. + +The `.to_native()` method is called to convert the initial datatype into a primative, serializable datatype. The `from_native()` method is called to restore a primative datatype into it's initial representation. + +## Examples + +Let's look at an example of serializing a class that represents an RGB color value: + + class Color(object): + """ + A color represented in the RGB colorspace. + """ + def __init__(self, red, green, blue): + assert(red >= 0 and green >= 0 and blue >= 0) + assert(red < 256 and green < 256 and blue < 256) + self.red, self.green, self.blue = red, green, blue + + class ColourField(serializers.WritableField): + """ + Color objects are serialized into "rgb(#, #, #)" notation. + """ + def to_native(self, obj): + return "rgb(%d, %d, %d)" % (obj.red, obj.green, obj.blue) + + def from_native(self, data): + data = data.strip('rgb(').rstrip(')') + red, green, blue = [int(col) for col in data.split(',')] + return Color(red, green, blue) + + +By default field values are treated as mapping to an attribute on the object. If you need to customize how the field value is accessed and set you need to override `.field_to_native()` and/or `.field_from_native()`. + +As an example, let's create a field that can be used represent the class name of the object being serialized: + + class ClassNameField(serializers.Field): + def field_to_native(self, obj, field_name): + """ + Serialize the object's class name. + """ + return obj.__class__ + [cite]: https://docs.djangoproject.com/en/dev/ref/forms/api/#django.forms.Form.cleaned_data [FILE_UPLOAD_HANDLERS]: https://docs.djangoproject.com/en/dev/ref/settings/#std:setting-FILE_UPLOAD_HANDLERS [strftime]: http://docs.python.org/2/library/datetime.html#strftime-and-strptime-behavior diff --git a/docs/api-guide/serializers.md b/docs/api-guide/serializers.md index ee7208a36..42e81cad5 100644 --- a/docs/api-guide/serializers.md +++ b/docs/api-guide/serializers.md @@ -87,27 +87,21 @@ By default, serializers must be passed values for all required fields or they wi serializer = CommentSerializer(comment, data={'content': u'foo bar'}, partial=True) # Update `instance` with partial data -## Serializing multiple objects - -To serialize a queryset or list of objects instead of a single object instance, you should pass the `many=True` flag when instantiating the serializer. - - queryset = Comment.objects.all() - serializer = CommentSerializer(queryset, many=True) - serializer.data - # [ - # {'email': u'leila@example.com', 'content': u'foo bar', 'created': datetime.datetime(2012, 8, 22, 16, 20, 9, 822774)}, - # {'email': u'jamie@example.com', 'content': u'baz', 'created': datetime.datetime(2013, 1, 12, 16, 12, 45, 104445)} - # ] - ## Validation -When deserializing data, you always need to call `is_valid()` before attempting to access the deserialized object. If any validation errors occur, the `.errors` property will contain a dictionary representing the resulting error messages. +When deserializing data, you always need to call `is_valid()` before attempting to access the deserialized object. If any validation errors occur, the `.errors` property will contain a dictionary representing the resulting error messages. For example: + + serializer = CommentSerializer(data={'email': 'foobar', 'content': 'baz'}) + serializer.is_valid() + # False + serializer.errors + # {'email': [u'Enter a valid e-mail address.'], 'created': [u'This field is required.']} Each key in the dictionary will be the field name, and the values will be lists of strings of any error messages corresponding to that field. The `non_field_errors` key may also be present, and will list any general validation errors. When deserializing a list of items, errors will be returned as a list of dictionaries representing each of the deserialized items. -### Field-level validation +#### Field-level validation You can specify custom field-level validation by adding `.validate_` methods to your `Serializer` subclass. These are analagous to `.clean_` methods on Django forms, but accept slightly different arguments. @@ -130,7 +124,7 @@ Your `validate_` methods should either just return the `attrs` dictio raise serializers.ValidationError("Blog post is not about Django") return attrs -### Object-level validation +#### Object-level validation To do any other validation that requires access to multiple fields, add a method called `.validate()` to your `Serializer` subclass. This method takes a single argument, which is the `attrs` dictionary. It should raise a `ValidationError` if necessary, or just return `attrs`. For example: @@ -151,26 +145,44 @@ To do any other validation that requires access to multiple fields, add a method ## Saving object state -Serializers also include a `.save()` method that you can override if you want to provide a method of persisting the state of a deserialized object. The default behavior of the method is to simply call `.save()` on the deserialized object instance. +To save the deserialized objects created by a serializer, call the `.save()` method: + + if serializer.is_valid(): + serializer.save() + +The default behavior of the method is to simply call `.save()` on the deserialized object instance. You can override the default save behaviour by overriding the `.save_object(obj)` method on the serializer class. The generic views provided by REST framework call the `.save()` method when updating or creating entities. ## Dealing with nested objects -The previous example is fine for dealing with objects that only have simple datatypes, but sometimes we also need to be able to represent more complex objects, -where some of the attributes of an object might not be simple datatypes such as strings, dates or integers. +The previous examples are fine for dealing with objects that only have simple datatypes, but sometimes we also need to be able to represent more complex objects, where some of the attributes of an object might not be simple datatypes such as strings, dates or integers. The `Serializer` class is itself a type of `Field`, and can be used to represent relationships where one object type is nested inside another. class UserSerializer(serializers.Serializer): - email = serializers.Field() - username = serializers.Field() + email = serializers.EmailField() + username = serializers.CharField(max_length=100) class CommentSerializer(serializers.Serializer): user = UserSerializer() - title = serializers.Field() - content = serializers.Field() - created = serializers.Field() + content = serializers.CharField(max_length=200) + created = serializers.DateTimeField() + +If a nested representation may optionally accept the `None` value you should pass the `required=False` flag to the nested serializer. + + class CommentSerializer(serializers.Serializer): + user = UserSerializer(required=False) # May be an anonymous user. + content = serializers.CharField(max_length=200) + created = serializers.DateTimeField() + +Similarly if a nested representation should be a list of items, you should the `many=True` flag to the nested serialized. + + class CommentSerializer(serializers.Serializer): + user = UserSerializer(required=False) + edits = EditItemSerializer(many=True) # A nested list of 'edit' items. + content = serializers.CharField(max_length=200) + created = serializers.DateTimeField() --- @@ -178,6 +190,96 @@ The `Serializer` class is itself a type of `Field`, and can be used to represent --- +## Dealing with multiple objects + +The `Serializer` class can also handle serializing or deserializing lists of objects. + +#### Serializing multiple objects + +To serialize a queryset or list of objects instead of a single object instance, you should pass the `many=True` flag when instantiating the serializer. You can then pass a queryset or list of objects to be serialized. + + queryset = Book.objects.all() + serializer = BookSerializer(queryset, many=True) + serializer.data + # [ + # {'id': 0, 'title': 'The electric kool-aid acid test', 'author': 'Tom Wolfe'}, + # {'id': 1, 'title': 'If this is a man', 'author': 'Primo Levi'}, + # {'id': 2, 'title': 'The wind-up bird chronicle', 'author': 'Haruki Murakami'} + # ] + +#### Deserializing multiple objects for creation + +To deserialize a list of object data, and create multiple object instances in a single pass, you should also set the `many=True` flag, and pass a list of data to be deserialized. + +This allows you to write views that create multiple items when a `POST` request is made. + +For example: + + data = [ + {'title': 'The bell jar', 'author': 'Sylvia Plath'}, + {'title': 'For whom the bell tolls', 'author': 'Ernest Hemingway'} + ] + serializer = BookSerializer(data=data, many=True) + serializer.is_valid() + # True + serializer.save() # `.save()` will be called on each deserialized instance + +#### Deserializing multiple objects for update + +You can also deserialize a list of objects as part of a bulk update of multiple existing items. +In this case you need to supply both an existing list or queryset of items, as well as a list of data to update those items with. + +This allows you to write views that update or create multiple items when a `PUT` request is made. + + # Capitalizing the titles of the books + queryset = Book.objects.all() + data = [ + {'id': 3, 'title': 'The Bell Jar', 'author': 'Sylvia Plath'}, + {'id': 4, 'title': 'For Whom the Bell Tolls', 'author': 'Ernest Hemingway'} + ] + serializer = BookSerializer(queryset, data=data, many=True) + serializer.is_valid() + # True + serialize.save() # `.save()` will be called on each updated or newly created instance. + +Bulk updates will update any instances that already exist, and create new instances for data items that do not have a corresponding instance. + +When performing a bulk update you may want any items that are not present in the incoming data to be deleted. To do so, pass `allow_delete=True` to the serializer. + + serializer = BookSerializer(queryset, data=data, many=True, allow_delete=True) + serializer.is_valid() + # True + serializer.save() # `.save()` will be called on each updated or newly created instance. + # `.delete()` will be called on any other items in the `queryset`. + +Passing `allow_delete=True` ensures that any update operations will completely overwrite the existing queryset, rather than simply updating any objects found in the incoming data. + +#### How identity is determined when performing bulk updates + +Performing a bulk update is slightly more complicated than performing a bulk creation, because the serializer needs a way of determining how the items in the incoming data should be matched against the existing object instances. + +By default the serializer class will use the `id` key on the incoming data to determine the canonical identity of an object. If you need to change this behavior you should override the `get_identity` method on the `Serializer` class. For example: + + class AccountSerializer(serializers.Serializer): + slug = serializers.CharField(max_length=100) + created = serializers.DateTimeField() + ... # Various other fields + + def get_identity(self, data): + """ + This hook is required for bulk update. + We need to override the default, to use the slug as the identity. + + Note that the data has not yet been validated at this point, + so we need to deal gracefully with incorrect datatypes. + """ + try: + return data.get('slug', None) + except AttributeError: + return None + +To map the incoming data items to their corresponding object instances, the `.get_identity()` method will be called both against the incoming data, and against the serialized representation of the existing objects. + ## Including extra context There are some cases where you need to provide extra context to the serializer in addition to the object being serialized. One common case is if you're using a serializer that includes hyperlinked relations, which requires the serializer to have access to the current request so that it can properly generate fully qualified URLs. @@ -190,47 +292,6 @@ You can provide arbitrary additional context by passing a `context` argument whe The context dictionary can be used within any serializer field logic, such as a custom `.to_native()` method, by accessing the `self.context` attribute. -## Creating custom fields - -If you want to create a custom field, you'll probably want to override either one or both of the `.to_native()` and `.from_native()` methods. These two methods are used to convert between the intial datatype, and a primative, serializable datatype. Primative datatypes may be any of a number, string, date/time/datetime or None. They may also be any list or dictionary like object that only contains other primative objects. - -The `.to_native()` method is called to convert the initial datatype into a primative, serializable datatype. The `from_native()` method is called to restore a primative datatype into it's initial representation. - -Let's look at an example of serializing a class that represents an RGB color value: - - class Color(object): - """ - A color represented in the RGB colorspace. - """ - def __init__(self, red, green, blue): - assert(red >= 0 and green >= 0 and blue >= 0) - assert(red < 256 and green < 256 and blue < 256) - self.red, self.green, self.blue = red, green, blue - - class ColourField(serializers.WritableField): - """ - Color objects are serialized into "rgb(#, #, #)" notation. - """ - def to_native(self, obj): - return "rgb(%d, %d, %d)" % (obj.red, obj.green, obj.blue) - - def from_native(self, data): - data = data.strip('rgb(').rstrip(')') - red, green, blue = [int(col) for col in data.split(',')] - return Color(red, green, blue) - - -By default field values are treated as mapping to an attribute on the object. If you need to customize how the field value is accessed and set you need to override `.field_to_native()` and/or `.field_from_native()`. - -As an example, let's create a field that can be used represent the class name of the object being serialized: - - class ClassNameField(serializers.Field): - def field_to_native(self, obj, field_name): - """ - Serialize the object's class name. - """ - return obj.__class__ - --- # ModelSerializers diff --git a/rest_framework/serializers.py b/rest_framework/serializers.py index 3029cf1c1..0b0c2c9cc 100644 --- a/rest_framework/serializers.py +++ b/rest_framework/serializers.py @@ -110,13 +110,15 @@ class BaseSerializer(Field): _dict_class = SortedDictWithMetadata def __init__(self, instance=None, data=None, files=None, - context=None, partial=False, many=None, source=None): + context=None, partial=False, many=None, source=None, + allow_delete=False): super(BaseSerializer, self).__init__(source=source) self.opts = self._options_class(self.Meta) self.parent = None self.root = None self.partial = partial self.many = many + self.allow_delete = allow_delete self.context = context or {} @@ -130,6 +132,12 @@ class BaseSerializer(Field): self._errors = None self._deleted = None + if many and instance is not None and not hasattr(instance, '__iter__'): + raise ValueError('instance should be a queryset or other iterable with many=True') + + if allow_delete and not many: + raise ValueError('allow_delete should only be used for bulk updates, but you have not set many=True') + ##### # Methods to determine which fields to use when (de)serializing objects. @@ -336,8 +344,15 @@ class BaseSerializer(Field): """ This hook is required for bulk update. It is used to determine the canonical identity of a given object. + + Note that the data has not been validated at this point, so we need + to make sure that we catch any cases of incorrect datatypes being + passed to this method. """ - return data.get('id') + try: + return data.get('id', None) + except AttributeError: + return None @property def errors(self): @@ -371,22 +386,12 @@ class BaseSerializer(Field): identities = [self.get_identity(self.to_native(obj)) for obj in objects] identity_to_objects = dict(zip(identities, objects)) - try: - iter(data) - if isinstance(data, dict): - raise TypeError - except TypeError: - self._errors = {'non_field_errors': ['Expected a list of items']} - else: + if hasattr(data, '__iter__') and not isinstance(data, (dict, six.text_type)): for item in data: if update: # Determine which object we're updating - try: - identity = self.get_identity(item) - except: - self.object = None - else: - self.object = identity_to_objects.pop(identity, None) + identity = self.get_identity(item) + self.object = identity_to_objects.pop(identity, None) ret.append(self.from_native(item, None)) errors.append(self._errors) @@ -395,6 +400,8 @@ class BaseSerializer(Field): self._deleted = identity_to_objects.values() self._errors = any(errors) and errors or [] + else: + self._errors = {'non_field_errors': ['Expected a list of items']} else: ret = self.from_native(data, files) @@ -445,7 +452,7 @@ class BaseSerializer(Field): else: self.save_object(self.object, **kwargs) - if self._deleted: + if self.allow_delete and self._deleted: [self.delete_object(item) for item in self._deleted] return self.object @@ -736,3 +743,13 @@ class HyperlinkedModelSerializer(ModelSerializer): 'many': to_many } return HyperlinkedRelatedField(**kwargs) + + def get_identity(self, data): + """ + This hook is required for bulk update. + We need to override the default, to use the url as the identity. + """ + try: + return data.get('url', None) + except AttributeError: + return None diff --git a/rest_framework/tests/serializer_bulk_update.py b/rest_framework/tests/serializer_bulk_update.py index 2f4164880..afc1a1a9f 100644 --- a/rest_framework/tests/serializer_bulk_update.py +++ b/rest_framework/tests/serializer_bulk_update.py @@ -201,7 +201,29 @@ class BulkUpdateSerializerTests(TestCase): 'author': 'Haruki Murakami' } ] - serializer = self.BookSerializer(self.books(), data=data, many=True) + serializer = self.BookSerializer(self.books(), data=data, many=True, allow_delete=True) + self.assertEqual(serializer.is_valid(), True) + self.assertEqual(serializer.data, data) + serializer.save() + new_data = self.BookSerializer(self.books(), many=True).data + self.assertEqual(data, new_data) + + def test_bulk_update_and_create(self): + """ + Bulk update serialization may also include created items. + """ + data = [ + { + 'id': 0, + 'title': 'The electric kool-aid acid test', + 'author': 'Tom Wolfe' + }, { + 'id': 3, + 'title': 'Kafka on the shore', + 'author': 'Haruki Murakami' + } + ] + serializer = self.BookSerializer(self.books(), data=data, many=True, allow_delete=True) self.assertEqual(serializer.is_valid(), True) self.assertEqual(serializer.data, data) serializer.save() @@ -227,6 +249,6 @@ class BulkUpdateSerializerTests(TestCase): {}, {'id': ['Enter a whole number.']} ] - serializer = self.BookSerializer(self.books(), data=data, many=True) + serializer = self.BookSerializer(self.books(), data=data, many=True, allow_delete=True) self.assertEqual(serializer.is_valid(), False) self.assertEqual(serializer.errors, expected_errors)