虚拟环境
Isolate 项目 dependencies 使用 venv, activate and deactivate, and manage 多个 Python versions.
你将学到
- 创建虚拟环境使用 venv
- Activate and deactivate 环境
- 理解为什么 isolation matters
- Use 需求.txt 到 pin dependencies
- Manage 多个 Python versions
概念讲解
Why Virtual Environments?
Without virtual environments, all your Python projects share the same global site-包 directory. If Project A needs Django 3 and Project B needs Django 4, they conflict. Virtual environments solve this by giving each project its own isolated Python environment with its own package directory.
A virtual environment is a self-contained directory containing:
- A Python interpreter (symlink to a specific version)
- Its own
site-包directory - Its own installed packages
- Activation scripts
Creating a Virtual Environment
Use the built-in venv module (Python 3.3+):
# Create a virtual environment named 'venv' (or '.venv')
python -m venv venv
# Or specify a Python version
python3.11 -m venv venv
This creates a venv/ directory containing the isolated Python environment. The common convention is to name it .venv (with a dot) so it is hidden in file explorers, or venv (without a dot).
Activating
Activation modifies your shell's PATH so the virtual environment's Python and pip are used first:
# Windows (Command Prompt)
venv\Scripts\activate
# Windows (PowerShell)
venv\Scripts\Activate.ps1
# macOS / Linux
source venv/bin/activate
After activation, your prompt changes to show the environment name:
(venv) $ python --version
(venv) $ pip install django
Now python and pip refer to the virtual environment's versions, and installed packages go into venv/lib/site-包/ instead of the global directory.
Deactivating
deactivate
This restores your shell to the global Python environment. The virtual environment still exists on disk — you can reactivate it anytime with the activate command.
requirements.txt
Pin your dependencies in a 需求.txt file so others can reproduce your environment:
# requirements.txt
django==4.2.7
requests>=2.31.0
numpy~=1.26.0
pytest==7.4.3
Install from it:
pip install -r requirements.txt
Generate it from your current environment:
pip freeze > requirements.txt
Pin exact versions (==) for reproducible builds. Use >= only if you intentionally want to allow upgrades. The ~= operator allows patch upgrades (e.g., ~=1.26.0 allows 1.26.x but not 1.27).
.gitignore
Never commit the virtual environment directory — it is platform-specific and large. Add it to .gitignore:
# .gitignore
venv/
.venv/
env/
__pycache__/
Commit 需求.txt instead — anyone can recreate the environment from it.
Multiple Python Versions
To manage multiple Python versions on one machine:
pyenv(macOS/Linux) — install and switch between Python versionspyenv-win(Windows) — Windows port of pyenv- Official Python installer — install multiple versions side by side
# With pyenv
pyenv install 3.11.7
pyenv install 3.12.1
pyenv local 3.11.7 # set version for current directory
# Create a venv with a specific version
python3.11 -m venv venv311
python3.12 -m venv venv312
Modern Alternatives
poetry— dependency management and packaging in one toolpipenv— Pipfile + Pipfile.lock for deterministic environmentsuv— ultra-fast Rust-based replacement for pip and venvconda— popular in data science, manages non-Python dependencies too
For beginners, venv + pip + 需求.txt is the standard starting point. Graduate to poetry or uv as your projects grow.
示例代码
# --- Creating and using a virtual environment ---
# These are shell commands, not Python code:
# 1. Create the virtual environment
# python -m venv .venv
# 2. Activate it
# Windows: .venv\Scripts\activate
# macOS: source .venv/bin/activate
# 3. Install packages (they go into .venv, not global)
# pip install django requests
# 4. Save dependencies
# pip freeze > requirements.txt
# 5. Deactivate when done
# deactivate
# --- requirements.txt example ---
# django==4.2.7
# requests>=2.31.0,<3.0.0
# numpy~=1.26.0
# pytest==7.4.3
# --- .gitignore example ---
# .venv/
# venv/
# __pycache__/
# *.pyc
# --- Python code to check which environment you're in ---
import sys
import os
# Check if running in a virtual environment
in_venv = hasattr(sys, 'real_prefix') or (
hasattr(sys, 'base_prefix') and sys.base_prefix != sys.prefix
)
print(f"Running in virtual environment: {in_venv}")
print(f"Python executable: {sys.executable}")
print(f"Prefix: {sys.prefix}")
# List installed packages
import subprocess
result = subprocess.run(
[sys.executable, "-m", "pip", "list"],
capture_output=True, text=True
)
print("Installed packages:")
print(result.stdout[:500]) # first 500 chars
Shows 完整 venv workflow: 创建, activate, 安装, freeze, deactivate. 包含需求.txt 示例, .gitignore 条目, and Python code 到 detect and inspect current 虚拟环境.
动手试一试
- JSON Formatter
Dependency files 如 pyproject.toml use structured 格式 — inspect 它们.
- 情况 Converter
包 and 环境 naming 约定.
常见错误
错误写法
Committing 虚拟环境 directory 到 version control
正确做法
venv directory 是 platform-特定, 大, and 不 portable. 添加 'venv/' and '.venv/' 到 .gitignore and 提交需求.txt 相反. Anyone 能 recreate 环境使用 'pip 安装 -r 需求.txt'.
错误写法
Forgetting 到 activate 虚拟环境之前安装包
正确做法
如果你 运行 pip 安装不使用 activating, 包 go 到全局 Python 环境, affecting 所有项目. 总是 activate 首先: 检查那个你的 prompt shows (venv) 之前安装. Use '哪个 python' 到验证你 是 using venv's Python.
相关资源
相关速查表
- Python Cheatsheet
venv commands, activation, and 需求.txt 语法.
查阅不认识的术语
新手友好的术语表用大白话解释编程术语——变量、函数、DOM、Promise 等等。
动手做项目
通过引导式项目应用所学,含起始代码与解决方案。
解读错误信息
用大白话解释常见错误——什么意思、为什么发生、如何修复。