mycroforge-docs/docs/tutorial.md
2024-10-06 13:24:20 +02:00

331 lines
11 KiB
Markdown

---
sidebar_position: 3
---
# Tutorial
In this tutorial, we'll build a simple todo app to demonstrate the capabilities of the MycroForge CLI.
By the end, you should have a solid foundation to start exploring and using MycroForge for your projects.
## General notes
The commands in this tutorial assume that you are running them from the root directory of your MycroForge project.
## Initialize the Project
Open a terminal and navigate (`cd`) to the directory where your project should be created.
Run the following command to initialize a new project and open it in VSCode:
```bash
m4g init todo-app
```
## Setup the database
Our todo app needs to keep track of todos, so it needs a storage mechanism of some sorts. A database should be one
of the first things, if not THE first thing, that comes to mind. Luckily, MycroForge provides you with a locally hosted
database for you project out of the box. This setup, is powered by docker compose and can be examined by opening the
`db.docker-compose.yml` file in the project. Follow along to learn how to use the docker based database when developing
locally.
### Run the database
The first step is to start the database, you can do this by running the following command in a terminal.
```bash
m4g db run
```
This command starts the services defined in the `db.docker-compose.yml` file.
You can verify that the services are up by running `docker container ls`. If everything went well, then the previous
command should output the service names defined in `db.docker-compose.yml`.
Go to [PhpMyAdmin (i.e. http://localhost:5051)](http://localhost:5051). You should now be able to login with the
following credentials.
- user: root
- pass: password
When you're done developing, you can shut down the local database by running `m4g db stop`
:::info
If you're running on MacOS, Docker might complain about a platform mismatch for PhpMyAdmin.
In that case you might need to specify the platform for the PhpMyAdmin image.
You can do this by passing the `--dbu-platform` flag to `m4g init`.
Run `m4g init -?` for all the available options.
If you've already initialized a project, you can also change the platform prefix of the PhpMyAdmin image in the `db.docker-compose.yml`.
:::
### Create the entities
Now that the database is running, we can start creating our entities. Run the commands below to create the `Todo` &
`Tag` entities.
```bash
m4g db generate entity Tag --column "description:str:String(255)"
```
```bash
m4g db generate entity Todo --column "description:str:String(255)" -c "is_done:bool:Boolean()"
```
After running these commands, you should find the generated entities in the `db/entities` folder of your project.
You should also see that the `main.py` & `db/env.py` files have been modified to include the newly generated entity.
For more information about the `m4g db generate entity` command, you can run `m4g db generate entity -?`.
### Define a many-to-many relation between Todo & Tag
To allow for relations between `Todo` & `Tag`, we'll define a many-to-many relation between the two entities.
This relation makes sense, because a `Todo` can have many `Tags` and a `Tag` could belong to many `Todos`.
You can generate this relation by running the following command from you terminal.
Creating a one-to-many relation would also make sense, but for the purpose of demonstration we're going to demonstrate
the many-to-many relation, because this one is the most complex, since it requires an additional mapping to be included
in the database schema.
```bash
m4g db link many Todo --to-many Tag
```
After running this command you should see that both the `Todo` and `Tag` entities now have a new field referencing the
a `List` containing instances of the other entity.
For more information about the `m4g db link` command try running `m4g db link -?`. Note that you can do the same thing
for all sub commands, so if you want to know more about `m4g db link many` you can simply run `m4g db link many -?` to
examine the command. The same is true for all the other commands as well.
### Generate the migration
Now that we've generated our entities, it's time to generate a migration that will apply these changes in the database.
Generate the initial migration by running the following command.
```bash
m4g db generate migration initial_migration
```
After running this command, you should see the new migration in the `db/version` directory.
### Apply the migration
The last step for the database setup is to actually apply the new migration to the database. This can be done by running
the following command.
```bash
m4g db migrate
```
After running this command, you should now see a populated schema when visiting [PhpMyAdmin](http://localhost:5051).
If for whatever reason you want to undo the last migration, you can simply run `m4g db rollback`.
## Setup the API
### Generate CRUD for Tag & Todo
Our API should provide use with basic endpoint to manage the `Todo` & `Tag` entities, i.e. CRUD functionality.
Writing this code can be boring, since it's pretty much boilerplate with some custom additions sprinkled here and there.
Fortunately, MycroForge can generate a good chunk of this boring code on your behalf. Run the following commands to
generate CRUD functionality for the `Todo` & `Tag` classes.
```bash
m4g api generate crud Tag
```
```bash
m4g api generate crud Todo
```
After running this command you should see that the `api/requests`,`api/routers` & `api/services` now contain the
relevant classes need to support the generated CRUD functionality. This could should be relatively straightforward, so
we won't dive into it, but feel free to take a break and explore what the generated code actually does. Another thing to
note, is that the generated routers are also automatically included in `main.py`.
### Modify the generated Todo request classes
Since we have a many-to-many relationship between `Todo` & `Tag`, the generated CRUD functionality isn't quite ready
yet. We need to be able to specify which `Tags` to add to a `Todo` when creating or updating it.
To do this, we will allow for a `tag_ids` field in both the `CreateTodoRequest` & the `UpdateTodoRequest`.
This field will contain the ids of the `Tags` that are associated with a `Todo`.
Modify `CreateTodoRequest` in `api/requests/create_todo_request.py`.
```python
# Before
from pydantic import BaseModel
class CreateTodoRequest(BaseModel):
description: str = None
is_done: bool = None
# After
from typing import List, Optional
from pydantic import BaseModel
class CreateTodoRequest(BaseModel):
description: str = None
is_done: bool = None
tag_ids: Optional[List[int]] = []
```
Modify `UpdateTodoRequest` in `api/requests/update_todo_request.py`, you might need to import `List` from `typing`.
```python
# Before
from pydantic import BaseModel
from typing import Optional
class UpdateTodoRequest(BaseModel):
description: Optional[str] = None
is_done: Optional[bool] = None
# After
from pydantic import BaseModel
from typing import List, Optional
class UpdateTodoRequest(BaseModel):
description: Optional[str] = None
is_done: Optional[bool] = None
tag_ids: Optional[List[int]] = []
```
### Modify generated TodoService
The `TodoService` will also need to be updated to accomodate the management of `tag_ids`.
Add the following imports in `api/services/todo_service.py`.
```python
from sqlalchemy.orm import selectinload
from db.entities.tag import Tag
```
Modify `TodoService.list`
```python
# Before
async def list(self) -> List[Todo]:
async with async_session() as session:
stmt = select(Todo)
results = (await session.scalars(stmt)).all()
return results
# After
async def list(self) -> List[Todo]:
async with async_session() as session:
stmt = select(Todo).options(selectinload(Todo.tags))
results = (await session.scalars(stmt)).all()
return results
```
Modify `TodoService.get_by_id`
```python
# Before
async def get_by_id(self, id: int) -> Optional[Todo]:
async with async_session() as session:
stmt = select(Todo).where(Todo.id == id)
result = (await session.scalars(stmt)).first()
return result
# After
async def get_by_id(self, id: int) -> Optional[Todo]:
async with async_session() as session:
stmt = select(Todo).where(Todo.id == id).options(selectinload(Todo.tags))
result = (await session.scalars(stmt)).first()
return result
```
Modify `TodoService.create`
```python
# Before
async def create(self, data: Dict[str, Any]) -> None:
async with async_session() as session:
entity = Todo(**data)
session.add(entity)
await session.commit()
# After
async def create(self, data: Dict[str, Any]) -> None:
tag_ids = []
if "tag_ids" in data.keys():
tag_ids = data["tag_ids"]
del data["tag_ids"]
async with async_session() as session:
entity = Todo(**data)
if len(tag_ids) > 0:
stmt = select(Tag).where(Tag.id.in_(tag_ids))
result = (await session.scalars(stmt)).all()
entity.tags = list(result)
session.add(entity)
await session.commit()
```
Modify `TodoService.update`
```python
# Before
async def update(self, id: int, data: Dict[str, Any]) -> bool:
tag_ids = []
if "tag_ids" in data.keys():
tag_ids = data["tag_ids"]
del data["tag_ids"]
async with async_session() as session:
stmt = select(Todo).where(Todo.id == id)
entity = (await session.scalars(stmt)).first()
if entity is None:
return False
else:
for key, value in data.items():
setattr(entity, key, value)
await session.commit()
return True
# After
async def update(self, id: int, data: Dict[str, Any]) -> bool:
tag_ids = []
if "tag_ids" in data.keys():
tag_ids = data["tag_ids"]
del data["tag_ids"]
async with async_session() as session:
stmt = select(Todo).where(Todo.id == id).options(selectinload(Todo.tags))
entity = (await session.scalars(stmt)).first()
if entity is None:
return False
else:
for key, value in data.items():
setattr(entity, key, value)
if len(tag_ids) > 0:
stmt = select(Tag).where(Tag.id.in_(tag_ids))
result = (await session.scalars(stmt)).all()
entity.tags = list(result)
else:
entity.tags = []
await session.commit()
return True
```
## Test the API!
Run the following command.
```bash
m4g api run
```
Go to http://localhost:5000/docs and test your Todo API!