CodeBlanks Logo Icon CodeBlanks
Login Register

setuptools

Learn packaging with setuptools.

setuptools

setuptools is a tool that builds Python packages and uploads them to PyPI. In modern usage it is not called directly; it is declared in the [build-system] table inside pyproject.toml.

[build-system]
requires = ["setuptools>=61"]
build-backend = "setuptools.build_meta"

requires lists the packages needed to build; build-backend names the backend to run. This tells pip install which tool to use.

The [build-system] table

The [build-system] table has two main fields. requires is the list of packages installed during build; build-backend is the dotted path to the Python object that drives the build.

[build-system]
requires = ["setuptools>=61"]
build-backend = "setuptools.build_meta"

The setuptools build_meta backend builds a wheel from source. Without this configuration pip install would not know which tool to call.

Practice Read the build-backend with tomllib

Start with import tomllib. Convert the TOML string below into structured data with tomllib.loads(...) and assign it to data:

[build-system]
requires = ["setuptools>=61"]
build-backend = "setuptools.build_meta"

Then print the backend path with print(data["build-system"]["build-backend"]).

Starter code

main.py
# bring in the tomllib module
# load the build-system table as structured data
# print the build-backend field

Sample solution

main.py
import tomllib

data = tomllib.loads('''
[build-system]
requires = ["setuptools>=61"]
build-backend = "setuptools.build_meta"
''')

print(data["build-system"]["build-backend"])

Expected output

setuptools.build_meta

Questions and answers

Review the lesson questions together with their answers. To test yourself, solve the lesson interactively.

1 Complete the build-backend line

Question 
[build-system]
requires = ["setuptools>=61"]
____ = "setuptools.build_meta"
Answer 
[build-system]
requires = ["setuptools>=61"]
build-backend = "setuptools.build_meta"

Explanation: build-backend names the build backend.

2 What does setuptools.build_meta do?

  • Uploads the package to PyPI
  • Builds a wheel from source Correct answer
  • Creates a virtual environment
  • Freezes dependencies

Explanation: build_meta builds a wheel from source.

3 Match the setuptools concepts

Term Meaning
setuptools Build tool
build_meta Produces a wheel
requires Build packages
wheel Installable artifact

All lessons in Advanced

1 Advanced decorators

  1. Parameterized decorator
  2. Class decorator
  3. functools
  4. functools.partial
  5. functools.reduce
  6. Summary - Advanced decorators

2 Type hints

  1. Basic hints
  2. Optional and Union
  3. typing.List and Dict
  4. Callable
  5. TypeVar
  6. Summary - Type hints

3 Pytest

  1. pytest basics
  2. assert
  3. fixture
  4. parametrize
  5. monkeypatch
  6. Summary - Pytest

4 Mocking

  1. unittest.mock
  2. patch
  3. MagicMock
  4. side_effect
  5. assert_called
  6. Summary - Mocking

5 Async basics

  1. async/await
  2. asyncio.run
  3. async def
  4. await asyncio.sleep
  5. create_task
  6. Summary - Async basics

6 Asyncio

  1. gather
  2. wait
  3. Queue
  4. Lock
  5. Semaphore
  6. Summary - Asyncio

7 HTTP and aiohttp

  1. urllib
  2. aiohttp.ClientSession
  3. async get
  4. async post
  5. JSON response
  6. Summary - HTTP and aiohttp

8 SQLAlchemy

  1. Engine
  2. Session
  3. Defining models
  4. Column and type
  5. Query
  6. Summary - SQLAlchemy

9 FastAPI

  1. FastAPI app
  2. @app.get
  3. Pydantic model
  4. Path parameter
  5. Query parameter
  6. Summary - FastAPI

10 Pydantic

  1. BaseModel
  2. Field
  3. validator
  4. Config
  5. parse_obj
  6. Summary - Pydantic

11 Design patterns

  1. Singleton
  2. Factory
  3. Strategy
  4. Observer
  5. Decorator
  6. Summary - Design patterns

12 Logging

  1. logging.getLogger
  2. Log levels
  3. basicConfig
  4. FileHandler
  5. dictConfig
  6. Summary - Logging

13 Packaging

  1. pyproject.toml
  2. setuptools
  3. Virtual environment
  4. pip install
  5. requirements.txt
  6. Summary - Packaging

14 Advanced final

  1. Project design
  2. Async layer
  3. SQLAlchemy layer
  4. FastAPI layer
  5. Putting it together