Dependency injection library
Project description
# FunDI
Solution for problem no one had before
Fun stays for function(or for fun if you wish) and DI for Dependency Injection
This library provides fast(to write!) and convenient(to use!) Dependency Injection for functional programming on python.
Why?
This library was inspired by FastAPI's dependency injection. The reasons for its existence are simple:
- A standalone dependency injection library. DI shouldn't be tied to a specific framework.
- It simplifies code writing. Dependency injection reduces boilerplate and improves maintainability.
- Lack of DI libraries for functional programming in Python. Or maybe I just didn't want to find one :3
How?
Main definitions
-
Scope - dependency injection scope (list of root values that will be set to parameters without dependency function)
Example:
from contextlib import ExitStack from fundi import inject, scan def dependant(value: str): print(value) with ExitStack() as stack: inject({"value": "Value that will be passed to dependency"}, scan(dependant), stack)
-
Dependant - Function that has dependencies (either function, or scope dependencies)
Example:
from contextlib import ExitStack from fundi import from_, inject, scan class Session: pass def string() -> str: return "string from dependency" def dependant(session: from_(Session), scope_value: str, dependency_value: str = from_(string)): pass with ExitStack() as stack: inject({"scope_value": "string from scope", "session": Session()}, scan(dependant), stack)
-
Dependency - Function that is used to resolve dependant's parameter value. Dependency can have its own dependencies.
Example:
from contextlib import ExitStack from fundi import from_, inject, scan class Session: pass def string() -> str: return "string from dependency" def dependency(session: from_(Session), scope_value: str, dependency_value: str = from_(string)): pass def dependant(value: None = from_(dependency)): pass with ExitStack() as stack: inject({"scope_value": "string from scope", "session": Session()}, scan(dependant), stack)
No more words, let's try!
Sync
from contextlib import ExitStack
from typing import Generator, Any
from fundi import from_, inject, scan
def require_database_session(database_url: str) -> Generator[str, Any]:
print(f"Opened database session at {database_url = }")
yield "database session"
print("Closed database session")
def require_user(session: str = from_(require_database_session)) -> str:
return "user"
def application(user: str = from_(require_user), session: str = from_(require_database_session)):
print(f"Application started with {user = }")
with ExitStack() as stack:
inject({"database_url": "postgresql://kuyugama:insecurepassword@localhost:5432/database"}, scan(application), stack)
Async
import asyncio
from contextlib import AsyncExitStack
from typing import AsyncGenerator, Any
from fundi import from_, ainject, scan
async def require_database_session(database_url: str) -> AsyncGenerator[str, Any]:
print(f"Opened database session at {database_url = }")
yield "database session"
print("Closed database session")
async def require_user(session: str = from_(require_database_session)) -> str:
return "user"
async def application(user: str = from_(require_user), session: str = from_(require_database_session)):
print(f"Application started with {user = }")
async def main():
async with AsyncExitStack() as stack:
await ainject({"database_url": "postgresql://kuyugama:insecurepassword@localhost:5432/database"}, scan(application), stack)
asyncio.run(main())
Resolve scope dependencies by type
It's simple! Use
from_on type annotation
from contextlib import ExitStack
from fundi import from_, inject, scan
class Session:
"""Database session"""
def require_user(_: from_(Session)) -> str:
return "user"
def application(session: from_(Session), user: str = from_(require_user)):
print(f"Application started with {user = }")
print(f"{session = }")
with ExitStack() as stack:
inject({"db": Session()}, scan(application), stack)
Note: while resolving dependencies by type - parameter names doesn't really matter.
Overriding dependency results
As simple, as injecting!
import time
import random
from contextlib import ExitStack
from fundi import from_, inject, scan
def require_random_animal() -> str:
time.sleep(3) # simulate web request
return random.choice(["cat", "dog", "chicken", "horse", "platypus", "cow"])
def application(animal: str = from_(require_random_animal)):
print("Animal:", animal)
with ExitStack() as stack:
inject({}, scan(application), stack, override={require_random_animal: "dog"})
Component explanation:
-
fundi.from_- Helps define dependencies.Use cases:
-
Tell resolver to resolve parameter value by its type, not name (ex:
parameter: from_(Session)) -
Define dependency (ex:
parameter: type = from_(dependency)).In this case it can be replaced with
fundi.scan.scan(ex:parameter: type = scan(dependency))
-
-
fundi.scan- Scans function for dependencies. ReturnsCallableInfoobject that is used by all functions that resolve dependencies. -
fundi.order- returns order in which dependencies will be resolved -
fundi.tree- returns dependency resolving tree
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file fundi-0.0.6.tar.gz.
File metadata
- Download URL: fundi-0.0.6.tar.gz
- Upload date:
- Size: 31.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.5.26
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c8eff79a01c2a76a82b6ebe81c875aca128736f89f1a9541c84aac7fb85042aa
|
|
| MD5 |
d0405952405553adedd28cf0b99eab84
|
|
| BLAKE2b-256 |
bbbbebb2d005dd4fe5919be197175d7db6cecc96c3cdd3ff848be537729b10e8
|
File details
Details for the file fundi-0.0.6-py3-none-any.whl.
File metadata
- Download URL: fundi-0.0.6-py3-none-any.whl
- Upload date:
- Size: 32.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.5.26
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8680639053868a06f75b07fc76affc526e7936654ed7ffa140af435f6b1fce93
|
|
| MD5 |
5c243ebbcc7f85e96b74a6ecc79f3b0f
|
|
| BLAKE2b-256 |
d7032c11756411da0615551fc247ffefdc32ff70fb49d0468ad7f2c9692ede2f
|