Description

Currently we have PypikaRepository supporting just SQLlite with promised support for other types (MySQL, PostgreSQL),
but to add this support we will need to create dedicated repository classes such as PostgreSqlRepository, because combining everything into one class will result in tons of conditions and unnecessary complication of code.

Possible Solutions

• Rename current version of PypikaRepository to SQLliteRepository, but can result in code duplication for generic pypika operations.
• Extract generic methods for pypika into PypikaRepository and create SQLliteRepository with SQLlite-related code

Description

Even though TinyDB is meant to be schemaless, we could have a tool like yoyo for sql to do changes in the existent objects to conform to the new schema.

Possible Solution

Create a TinyDB extension that can read a list of files from a migrations directory and execute the steps defined there. Each file will have a hash, and the latest applied can be saved in the _tinydb_migrations table in the database.

I'd copy how yoyo has already solved the issue.

Description

A repository that uses GraphQL as backend.

Possible Solution

Maybe we could use Strawberry-graphql

Additional context

Related Issue

Description

Create a repository with a mongoDB backend

Possible Solution

Maybe we could use OMDantic

Additional context

Related Issue

First of all thanks for the very useful library. Here's my question:
I use an "Entity" class like this:

class MyClass(Entity):
    foo: str
    ...

I want to add a list of MyClass objects like this:

repo.add(my_objs)

my_objs has the type List[MyClass]
This works perfectly but the problem is that surprisingly I get a mypy error:

Value of type variable "EntityOrEntities" of "add" of "Repository" cannot be "List[MyClass]"

Am I doing something wrong? How can I resolve this?

Thank you very much

Test that the PypikaRepository supports PostgreSQL databases

Description

It would be nice if the repositories could load a schema to understand the relationship between entities.

In this schema we could define the following relationships:

• One to one
• One to many
• Many to many

In a way that when we .get an object we get the option to load the mappers, and if they are, the repo will get the required objects and load them in the specified attribute.

Possible Solution

At repo initialization, a schema attribute is set with the desired relationships.

Then we will need to decide how it affects the different base methods.

Additional context

Related Issue