CODEWITHSUNDEEP
python-3.xgoogle-app-enginegoogle-cloud-datastore
daiyue
daiyue

Reputation: 7448

google datastore ValueError: Only a partial key can be completed

I tried to add a collection of entities into datastore using a list of dictionaries, the code is as follows,

    def add_rows(self, val_dicts):
        with self.client.transaction():
            entities = [Entity(self.client.key(self.kind))] * len(val_dicts)

            for entity, update_dict in zip(entities, val_dicts):
                entity.update(update_dict)

            self.client.put_multi(entities)

when I ran the code, I got the following error,

ValueError: Only a partial key can be completed.

But if I changed the code to,

    def add_rows(self, val_dicts):
        with self.client.transaction():

            entities = [Entity(self.client.key(self.kind)) for i in range(len(val_dicts))]

            for entity, update_dict in zip(entities, val_dicts):
                entity.update(update_dict)

            self.client.put_multi(entities)

the error is gone. But there is no difference in creating the entities, so I am wondering what is the issue in the first code snippet.

I am also using datastore emulator for testing these code.

Upvotes: 1

Views: 304

Answers (1)

Dan Cornilescu
Dan Cornilescu

Reputation: 39824

In your first code snippet you only create a single entity and entities is a list containing identical references to that single entity - because of the way the list is initialized. Which is likely the root cause of your problem - you're overwriting that entity with each val_dicts value and put_multi() barfs because it's called with the same entity multiple times as arguments.

In the second snippet each entities element is a diferent entity, no problem.

To illustrate:

>>> class Foo(object):
...  pass
... 
>>> [Foo()]*2
[<__main__.Foo object at 0x7fc9760eee80>, <__main__.Foo object at 0x7fc9760eee80>]
>>> [Foo() for i in range(2)]
[<__main__.Foo object at 0x7fc9760eeeb8>, <__main__.Foo object at 0x7fc9760eefd0>]
>>>

Note the identical object addresses in the 1st case.

Upvotes: 2

Related Questions