add JSONFileChildSeeder class

[liviomendonca]

We propose an extension of JSONFileModelSeeder which can handle child models. It has been implemented with a caching system to speed up the seeding process.

Also correct some minor typos.

[suliman-99]

First of all i should thank both of you for your contributions, i am so happy to see your commits
It is really really nice work and i like it

But I think there is redundancy in the code in this section
I find out that these two sections do the same thing and we can delete one of them
both of them get the needed object of a specific model with specific query data (dict) then store it in the hashmap and get it from it if it is requested again instead of getting it again from the database

first one between line 467 and line 477

                        if not related_keys_dict[k] in hashmap:
                            hashmap[related_keys_dict[k]] = dict()

                        tuple_v = _convert_to_tuple(v)

                        if tuple_v in hashmap[related_keys_dict[k]]:
                            entry[k] = hashmap[related_keys_dict[k]][tuple_v]

                        else:
                            entry[k] = _replace_item(v, related_keys_dict[k])
                            hashmap[related_keys_dict[k]][tuple_v] = entry[k]

second part between line 482 and line 491:

            obj_pk = _convert_to_tuple(entry)

            if not model in hashmap:
                hashmap[model] = dict()

            if obj_pk in hashmap[model]:
                return hashmap[model][obj_pk]

            hashmap[model][obj_pk] = model.objects.get(**entry)
            return hashmap[model][obj_pk]

maybe i misunderstand the code
correct me if i am wrong
and thank you in advance

[liviomendonca]

Thanks for the review!

You're absolutely correct! We changed our code to remove the redundancy ;)