Bài 14: Virtual Environments Deep Dive

Mục Tiêu Bài Học

Sau khi hoàn thành bài này, bạn sẽ:

✅ Hiểu venv vs virtualenv vs conda
✅ Sử dụng pyenv cho multiple Python versions
✅ Làm việc với pipenv và Pipfile
✅ Sử dụng poetry dependency management
✅ Áp dụng Docker containers

Tại Sao Cần Virtual Environments?

Virtual environments giải quyết các vấn đề:

1. Dependency Conflicts

# Project A cần Django 3.2pip install Django==3.2 # Project B cần Django 4.0pip install Django==4.0  # ❌ Conflict!

2. System Python Pollution

# Cài packages vào system Pythonsudo pip install requests  # ❌ Không nên làm!# Có thể làm hỏng system tools

3. Reproducible Environments

# Dễ dàng chia sẻ dependenciespip freeze > requirements.txtpip install -r requirements.txt

1. venv - Built-in Virtual Environments

venv là module có sẵn từ Python 3.3+.

Tạo và Kích Hoạt venv

# Tạo virtual environmentpython -m venv myenv # Kích hoạt (macOS/Linux)source myenv/bin/activate # Kích hoạt (Windows)myenv\Scripts\activate # Kiểm tra Python locationwhich python  # myenv/bin/python # Thoátdeactivate

Cài Đặt Packages

# Sau khi activatepip install requests pandas numpy # List installed packagespip list # Freeze dependenciespip freeze > requirements.txt # Install từ requirementspip install -r requirements.txt

Tự Động Hóa venv

setup_env.sh

#!/bin/bash # Tạo venv nếu chưa cóif [ ! -d "venv" ]; then    python3 -m venv venv    echo "✅ Created virtual environment"fi # Kích hoạtsource venv/bin/activate # Cài đặt dependenciespip install -r requirements.txt echo "✅ Environment ready!"

2. virtualenv - Advanced Virtual Environments

virtualenv mạnh hơn venv với nhiều tính năng.

Cài Đặt và Sử Dụng

# Cài virtualenvpip install virtualenv # Tạo virtualenvvirtualenv myenv # Tạo với Python version cụ thểvirtualenv -p python3.9 myenv # Không copy packages từ systemvirtualenv --no-site-packages myenv # Tạo với system site packagesvirtualenv --system-site-packages myenv

virtualenvwrapper

Tool quản lý nhiều virtualenvs:

# Cài đặtpip install virtualenvwrapper # Thêm vào ~/.bashrc hoặc ~/.zshrcexport WORKON_HOME=$HOME/.virtualenvssource /usr/local/bin/virtualenvwrapper.sh # Tạo virtualenvmkvirtualenv myproject # List tất cả virtualenvslsvirtualenv # Chuyển đổi virtualenvworkon myproject # Xóa virtualenvrmvirtualenv myproject # Copy virtualenvcpvirtualenv old_env new_env

3. pyenv - Python Version Management

pyenv quản lý nhiều Python versions.

Cài Đặt pyenv

# macOS (với Homebrew)brew install pyenv # Linuxcurl https://pyenv.run | bash # Thêm vào ~/.bashrc hoặc ~/.zshrcexport PATH="$HOME/.pyenv/bin:$PATH"eval "$(pyenv init -)"eval "$(pyenv virtualenv-init -)"

Sử Dụng pyenv

# List Python versions có sẵnpyenv install --list # Cài Python versionpyenv install 3.9.7pyenv install 3.10.4 # List installed versionspyenv versions # Set global Python versionpyenv global 3.10.4 # Set local Python version (cho project)pyenv local 3.9.7 # Set shell Python version (session hiện tại)pyenv shell 3.10.4

pyenv-virtualenv

# Tạo virtualenv với pyenvpyenv virtualenv 3.9.7 myproject-env # Kích hoạtpyenv activate myproject-env # Deactivatepyenv deactivate # Xóapyenv uninstall myproject-env # Auto-activate khi cd vào projectecho "myproject-env" > .python-version

4. pipenv - Modern Dependency Management

pipenv kết hợp virtualenv và dependency management.

Cài Đặt và Sử Dụng

# Cài pipenvpip install pipenv # Tạo project với pipenvcd myprojectpipenv install  # Tạo Pipfile và virtualenv # Cài packagespipenv install requestspipenv install pytest --dev  # Dev dependency # Kích hoạt shellpipenv shell # Run command trong virtualenvpipenv run python script.pypipenv run pytest # Lock dependenciespipenv lock # Install từ Pipfile.lockpipenv install --deploy

Pipfile Example

Pipfile

[[source]]url = "https://pypi.org/simple"verify_ssl = truename = "pypi" [packages]django = ">=4.0,<5.0"requests = "*"psycopg2-binary = "==2.9.3" [dev-packages]pytest = "*"black = "*"mypy = "*" [requires]python_version = "3.10" [scripts]dev = "python manage.py runserver"test = "pytest tests/"format = "black ."

pipenv Best Practices

# Chỉ commit Pipfile và Pipfile.lockgit add Pipfile Pipfile.lockgit commit -m "Add dependencies" # Trên productionpipenv install --deploy --ignore-pipfile # Update tất cả dependenciespipenv update # Check security vulnerabilitiespipenv check # Graph dependenciespipenv graph # Remove unused packagespipenv clean

5. Poetry - Advanced Dependency Management

poetry là tool hiện đại nhất cho dependency management.

Cài Đặt Poetry

# macOS/Linuxcurl -sSL https://install.python-poetry.org | python3 - # Windows PowerShell(Invoke-WebRequest -Uri https://install.python-poetry.org -UseBasicParsing).Content | python - # Kiểm trapoetry --version

Tạo Project với Poetry

# Tạo project mớipoetry new myproject # Structuremyproject/├── pyproject.toml├── README.md├── myproject/│   └── __init__.py└── tests/    └── __init__.py # Hoặc init trong existing projectcd existing_projectpoetry init

Quản Lý Dependencies với Poetry

# Thêm dependencypoetry add requestspoetry add pytest --group dev # Xóa dependencypoetry remove requests # Update dependenciespoetry update # Install tất cả dependenciespoetry install # Install không có dev dependenciespoetry install --only main # Show dependenciespoetry showpoetry show --tree

pyproject.toml Example

pyproject.toml

[tool.poetry]name = "myproject"version = "0.1.0"description = "My awesome project"authors = ["Your Name <you@example.com>"]readme = "README.md"packages = [{include = "myproject"}] [tool.poetry.dependencies]python = "^3.9"django = "^4.0"requests = "^2.28"psycopg2-binary = "^2.9" [tool.poetry.group.dev.dependencies]pytest = "^7.0"black = "^22.0"mypy = "^0.950"flake8 = "^4.0" [tool.poetry.scripts]dev = "myproject.cli:dev"test = "pytest" [build-system]requires = ["poetry-core"]build-backend = "poetry.core.masonry.api"

Poetry Virtual Environments

# Poetry tự động tạo virtualenvpoetry install  # Tạo venv và install deps # Run command trong venvpoetry run python script.pypoetry run pytest # Activate shellpoetry shell # Kiểm tra venv locationpoetry env info # List tất cả venvspoetry env list # Xóa venvpoetry env remove python3.9 # Sử dụng system Pythonpoetry config virtualenvs.create false

5 Ứng Dụng Thực Tế

1. Multi-Project Development Environment

project_manager.py

import subprocessimport osfrom pathlib import Path class ProjectManager:    """Quản lý nhiều projects với different Python versions"""        def __init__(self, workspace: str):        self.workspace = Path(workspace)        def create_project(self, name: str, python_version: str = "3.10"):        """Tạo project với Poetry và pyenv"""        project_path = self.workspace / name        project_path.mkdir(exist_ok=True)                os.chdir(project_path)                # Set Python version với pyenv        subprocess.run(["pyenv", "local", python_version])                # Tạo project với Poetry        subprocess.run(["poetry", "init", "--no-interaction"])                print(f"✅ Created {name} with Python {python_version}")        def list_projects(self):        """List tất cả projects và Python versions"""        for project in self.workspace.iterdir():            if project.is_dir():                version_file = project / ".python-version"                version = "unknown"                if version_file.exists():                    version = version_file.read_text().strip()                                print(f"📦 {project.name}: Python {version}")        def activate_project(self, name: str):        """Generate activation script"""        project_path = self.workspace / name                script = f"""#!/bin/bashcd {project_path}pyenv activatepoetry shell"""                script_path = project_path / "activate.sh"        script_path.write_text(script)        script_path.chmod(0o755)                print(f"✅ Run: {script_path}") # Sử dụngmanager = ProjectManager("/Users/dev/projects")manager.create_project("api-service", "3.10")manager.create_project("ml-pipeline", "3.9")manager.list_projects()

2. CI/CD Environment Manager

ci_env.py

import yamlfrom typing import Dict, List class CIEnvironment:    """Quản lý environments cho CI/CD"""        def __init__(self, project_path: str):        self.project_path = project_path        def generate_github_actions(self, python_versions: List[str]):        """Generate GitHub Actions workflow"""        workflow = {            "name": "Tests",            "on": ["push", "pull_request"],            "jobs": {                "test": {                    "runs-on": "ubuntu-latest",                    "strategy": {                        "matrix": {                            "python-version": python_versions                        }                    },                    "steps": [                        {                            "uses": "actions/checkout@v3"                        },                        {                            "name": "Set up Python",                            "uses": "actions/setup-python@v4",                            "with": {                                "python-version": "${{ matrix.python-version }}"                            }                        },                        {                            "name": "Install Poetry",                            "run": "pip install poetry"                        },                        {                            "name": "Install dependencies",                            "run": "poetry install"                        },                        {                            "name": "Run tests",                            "run": "poetry run pytest"                        }                    ]                }            }        }                path = f"{self.project_path}/.github/workflows/test.yml"        with open(path, "w") as f:            yaml.dump(workflow, f, default_flow_style=False)                print(f"✅ Generated GitHub Actions workflow")        def generate_dockerfile(self, python_version: str):        """Generate Dockerfile với Poetry"""        dockerfile = f"""FROM python:{python_version}-slim WORKDIR /app # Install PoetryRUN pip install poetry # Copy dependency filesCOPY pyproject.toml poetry.lock ./ # Install dependenciesRUN poetry config virtualenvs.create false \\    && poetry install --only main --no-interaction # Copy applicationCOPY . . CMD ["poetry", "run", "python", "main.py"]"""                path = f"{self.project_path}/Dockerfile"        with open(path, "w") as f:            f.write(dockerfile)                print(f"✅ Generated Dockerfile") # Sử dụngci = CIEnvironment("./myproject")ci.generate_github_actions(["3.9", "3.10", "3.11"])ci.generate_dockerfile("3.10")

3. Dependency Security Scanner

security_checker.py

import subprocessimport jsonfrom typing import List, Dict class DependencyScanner:    """Scan dependencies cho security vulnerabilities"""        def __init__(self, project_path: str):        self.project_path = project_path        def check_pipenv(self) -> List[Dict]:        """Check Pipenv dependencies"""        result = subprocess.run(            ["pipenv", "check", "--json"],            cwd=self.project_path,            capture_output=True,            text=True        )                if result.returncode != 0:            data = json.loads(result.stdout)            return data.get("vulnerabilities", [])                return []        def check_poetry(self) -> List[Dict]:        """Check Poetry dependencies với safety"""        # Install safety        subprocess.run(["pip", "install", "safety"])                # Export requirements        subprocess.run(            ["poetry", "export", "-f", "requirements.txt",              "-o", "requirements.txt"],            cwd=self.project_path        )                # Check với safety        result = subprocess.run(            ["safety", "check", "--json",              "-r", "requirements.txt"],            cwd=self.project_path,            capture_output=True,            text=True        )                if result.returncode != 0:            return json.loads(result.stdout)                return []        def generate_report(self):        """Generate security report"""        vulnerabilities = self.check_poetry()                if not vulnerabilities:            print("✅ No vulnerabilities found!")            return                print(f"⚠️  Found {len(vulnerabilities)} vulnerabilities:")                for vuln in vulnerabilities:            print(f"""Package: {vuln['package']}Installed: {vuln['installed_version']}Affected: {vuln['vulnerable_spec']}Description: {vuln['advisory']}Severity: {vuln['severity']}Fix: Upgrade to {vuln['analyzed_version']}""") # Sử dụngscanner = DependencyScanner("./myproject")scanner.generate_report()

4. Environment Synchronization Tool

env_sync.py

import subprocessimport jsonfrom pathlib import Pathfrom typing import Dict, Set class EnvironmentSync:    """Đồng bộ environments giữa các máy"""        def __init__(self, project_path: str):        self.project_path = Path(project_path)        def export_environment(self) -> Dict:        """Export environment configuration"""        config = {}                # Python version từ pyenv        version_file = self.project_path / ".python-version"        if version_file.exists():            config["python_version"] = version_file.read_text().strip()                # Dependencies từ Poetry        result = subprocess.run(            ["poetry", "export", "-f", "requirements.txt"],            cwd=self.project_path,            capture_output=True,            text=True        )        config["dependencies"] = result.stdout                # Environment variables        env_file = self.project_path / ".env"        if env_file.exists():            config["env_vars"] = env_file.read_text()                # Save config        config_file = self.project_path / "env_config.json"        config_file.write_text(json.dumps(config, indent=2))                print("✅ Exported environment configuration")        return config        def import_environment(self, config_path: str):        """Import và setup environment"""        config = json.loads(Path(config_path).read_text())                # Setup Python version        if "python_version" in config:            subprocess.run([                "pyenv", "install", "-s", config["python_version"]            ])            subprocess.run([                "pyenv", "local", config["python_version"]            ], cwd=self.project_path)                # Setup Poetry        subprocess.run(["poetry", "install"], cwd=self.project_path)                # Restore environment variables        if "env_vars" in config:            env_file = self.project_path / ".env"            env_file.write_text(config["env_vars"])                print("✅ Imported environment configuration")        def compare_environments(self, other_project: str) -> Dict:        """So sánh dependencies giữa 2 projects"""        def get_packages(path: Path) -> Set[str]:            result = subprocess.run(                ["poetry", "show"],                cwd=path,                capture_output=True,                text=True            )            return set(line.split()[0] for line in result.stdout.splitlines())                current = get_packages(self.project_path)        other = get_packages(Path(other_project))                return {            "only_in_current": current - other,            "only_in_other": other - current,            "common": current & other        } # Sử dụngsync = EnvironmentSync("./project-a")config = sync.export_environment() # Trên máy khácsync2 = EnvironmentSync("./project-a")sync2.import_environment("env_config.json") # So sánhdiff = sync.compare_environments("./project-b")print(f"Common packages: {len(diff['common'])}")

5. Docker Development Environment

docker_env.py

from typing import List, Dict class DockerEnvironment:    """Quản lý Docker development environments"""        def __init__(self, project_name: str):        self.project_name = project_name        def generate_dockerfile_multistage(self, python_version: str):        """Generate multi-stage Dockerfile"""        dockerfile = f"""# Stage 1: BuilderFROM python:{python_version} as builder WORKDIR /app # Install PoetryRUN pip install poetry # Copy dependency filesCOPY pyproject.toml poetry.lock ./ # Install dependenciesRUN poetry export -f requirements.txt -o requirements.txt \\    && pip wheel --no-cache-dir --no-deps --wheel-dir /app/wheels -r requirements.txt # Stage 2: RuntimeFROM python:{python_version}-slim WORKDIR /app # Copy wheels từ builderCOPY --from=builder /app/wheels /wheelsCOPY --from=builder /app/requirements.txt . # Install dependenciesRUN pip install --no-cache /wheels/* # Copy applicationCOPY . . # Run as non-root userRUN useradd -m appuser && chown -R appuser:appuser /appUSER appuser CMD ["python", "main.py"]"""                with open("Dockerfile", "w") as f:            f.write(dockerfile)                print("✅ Generated multi-stage Dockerfile")        def generate_docker_compose(self, services: List[Dict]):        """Generate docker-compose.yml"""        compose = {            "version": "3.8",            "services": {}        }                for service in services:            compose["services"][service["name"]] = {                "build": ".",                "volumes": [                    f"./{service['name']}:/app"                ],                "environment": service.get("env", {}),                "ports": service.get("ports", []),                "depends_on": service.get("depends_on", [])            }                # Add development service        compose["services"]["dev"] = {            "build": ".",            "volumes": [".:/app"],            "command": "poetry run python -m debugpy --listen 0.0.0.0:5678 main.py",            "ports": ["5678:5678"],            "stdin_open": True,            "tty": True        }                import yaml        with open("docker-compose.yml", "w") as f:            yaml.dump(compose, f, default_flow_style=False)                print("✅ Generated docker-compose.yml") # Sử dụngdocker_env = DockerEnvironment("myproject")docker_env.generate_dockerfile_multistage("3.10")docker_env.generate_docker_compose([    {        "name": "app",        "ports": ["8000:8000"],        "env": {"DEBUG": "1"}    }])

Best Practices

1. Project Structure

myproject/├── .python-version      # pyenv version├── pyproject.toml       # Poetry config├── poetry.lock          # Locked dependencies├── .env.example         # Environment variables template├── .gitignore           # Git ignore (include venv/, .env)├── README.md            # Setup instructions├── src/│   └── myproject/│       └── __init__.py└── tests/    └── test_*.py

2. Dependencies Management

# ✅ Pin important versionsDjango==4.0.4 # ✅ Allow patch updatesrequests~=2.28.0 # ✅ Allow minor updates (Poetry)django = "^4.0" # ❌ Avoid wildcardpackage = "*"

3. Environment Variables

# .env.example - Commit to gitDATABASE_URL=postgresql://localhost/mydbSECRET_KEY=change-me-in-productionDEBUG=True # .env - Don't commit!DATABASE_URL=postgresql://user:pass@localhost/mydbSECRET_KEY=actual-secret-keyDEBUG=False

4. Documentation

README.md

## Setup Development Environment ### Prerequisites- Python 3.10+- Poetry ### Installation1. Clone repository2. Install dependencies:   \`\`\`bash   poetry install   \`\`\`3. Copy environment variables:   \`\`\`bash   cp .env.example .env   \`\`\`4. Run application:   \`\`\`bash   poetry run python main.py   \`\`\`

5. Tool Selection Guide

Tool	Use Case
`venv`	Simple projects, learning
`virtualenv`	More features than venv
`pyenv`	Multiple Python versions
`pipenv`	Simple dependency management
`poetry`	Modern projects, publishing packages
`conda`	Data science, scientific computing
`Docker`	Production, microservices

Bài Tập Thực Hành

Bài 1: Setup Multi-Version Environment

Tạo 3 projects với Python 3.9, 3.10, 3.11 sử dụng pyenv và Poetry.

Bài 2: Migrate từ requirements.txt sang Poetry

Migrate một existing project từ requirements.txt sang Poetry với pyproject.toml.

Bài 3: CI/CD Pipeline

Tạo GitHub Actions workflow test code trên Python 3.9, 3.10, 3.11.

Bài 4: Docker Development Environment

Tạo Dockerfile và docker-compose.yml cho development với hot-reload.

Bài 5: Dependency Audit

Scan một project cho security vulnerabilities và update dependencies.

Tóm Tắt

Trong bài này chúng ta đã học:

✅ venv - Built-in virtual environments
✅ virtualenv - Advanced virtual environments với nhiều features
✅ pyenv - Quản lý multiple Python versions
✅ pipenv - Modern dependency management với Pipfile
✅ Poetry - Advanced dependency management và packaging
✅ Docker - Container-based development environments
✅ Best Practices - Project structure, dependencies, CI/CD

Virtual environments là foundation của Python development, giúp isolate dependencies và maintain reproducible environments.

Bài tiếp theo: Bài 15: Package Structure & Distribution - Học cách structure và distribute Python packages! 📦

#pipenv

Virtual Environments Deep Dive