Where to find out more about the MiniDataApi spec?

[soumendra]

• I have tried Google, no useful results
• I have tried fastlite repo, no mention

MiniDataApi spec sounds like something that'll be useful. Where can I find more about it?

[jph00]

Heh I gotta write it still! Coming soon :D

[anddam]

Heh I gotta write it still!

But this seems like something already existing

it'll be easier if you pick a lib that supports the MiniDataAPI spec.
Here we are using SQLite, with the FastLite library, which supports the MiniDataAPI spec.`

How are DBs or libraries supporting a spec that hasn't been written yet?

What are comments like

  # `xtra` is part of the MiniDataAPI spec. 

in adv_app.py or pep8_app.py referencing?

[jph00]

I wrote the lib first based on what's in my head around the API.

adv_app.py is the best file to look at.

[pigeonflight]

This is ChatGPT's best guess (I had it read the adv_app.py file)

Introduction to MiniData API Spec

The MiniData API spec is a lightweight database interaction standard designed to simplify and streamline the management of data within applications. It supports essential CRUD (Create, Read, Update, Delete) operations, schema management, and filtering, ensuring a flexible and efficient way to handle data. Below is a guide on how the MiniData API spec works, illustrated through an example application.

Basic Concepts

1. Database Initialization: The MiniData API spec allows for easy database initialization using libraries like SQLite. You can connect to a database and access its tables using a simple and consistent interface.

2. Table Management: Tables in the MiniData API are managed through a collection attribute (e.g., db.t). You can check for the existence of tables and create them if necessary, defining their schema directly in the code.

3. Dataclass Integration: The spec supports Python's dataclass to provide type safety and structure to the data objects returned from queries.

Key Features

1. Creating and Managing Tables:

   db = database('data/utodos.db')
   todos, users = db.t.todos, db.t.users
   
   if todos not in db.t:
       users.create(dict(name=str, pwd=str), pk='name')
       todos.create(id=int, title=str, done=bool, name=str, details=str, priority=int, pk='id')
   

2. Defining Data Models with Dataclasses:

   Todo, User = todos.dataclass(), users.dataclass()
   

3. CRUD Operations:

   • Create: Insert new records into the table.

     new_user = users.insert(dict(name='Alice', pwd='password123'))
     

   • Read: Query records from the table using primary keys or filters.

     user = users['Alice']  # Fetch by primary key
     todo_list = todos(name='Alice')  # Fetch with filter
     

   • Update: Modify existing records in the table.

     todos.update({'done': True}, 1)  # Mark todo with id 1 as done
     

   • Delete: Remove records from the table.

     todos.delete(1)  # Delete todo with id 1
     

4. Filtering and Security with xtra: The xtra method ensures that users can only access their own records by adding filters automatically.

   def before(req, sess):
       auth = req.scope['auth'] = sess.get('auth', None)
       if not auth: return RedirectResponse('/login', status_code=303)
       todos.xtra(name=auth)
   

Example Application Walkthrough

1. Setup and Initialization:

   • Initialize the database and tables.
   • Define data models using dataclasses.

   from fasthtml.common import *
   db = database('data/utodos.db')
   todos, users = db.t.todos, db.t.users
   
   if todos not in db.t:
       users.create(dict(name=str, pwd=str), pk='name')
       todos.create(id=int, title=str, done=bool, name=str, details=str, priority=int, pk='id')
   
   Todo, User = todos.dataclass(), users.dataclass()
   

2. CRUD Operations in Routes:

   • Example of a POST request handling user login.

   @rt("/login")
   def post(login:Login, sess):
       try: u = users[login.name]
       except NotFoundError: u = users.insert(login)
       if not compare_digest(u.pwd.encode("utf-8"), login.pwd.encode("utf-8")): return login_redir
       sess['auth'] = u.name
       return RedirectResponse('/', status_code=303)
   

3. Filtering with xtra:

   • Ensure users only see their own todos.

   def before(req, sess):
       auth = req.scope['auth'] = sess.get('auth', None)
       if not auth: return RedirectResponse('/login', status_code=303)
       todos.xtra(name=auth)
   

Conclusion

The MiniData API spec provides a simple and consistent way to manage database operations in Python applications. It supports essential CRUD functionalities, schema management, and data filtering, making it a powerful tool for developers to build robust and secure applications. By following the principles outlined in this guide, you can effectively utilize the MiniData API spec in your projects.

[anddam]

This is ChatGPT's best guess (I had it read the adv_app.py file)

Oh that's actually pretty good, thx.

[soumendra]

Can't wait for this to land

[pydanny]

The current draft is here. It needs some work, there's some missing items. I hope to have the first iteration in today!

[pydanny]

The MiniData spec is here! Many thanks to everyone who helped get it out the door. 😄