https://pyloong.github.io/pythonic-project-guidelines/practices/web/#39

开发环境搭建

安装虚拟环境工具:poetry

brew install pipx
pipx install poetry
pipx ensurepath

pipx upgrade poetry(需要时)

项目初始化

初始化项目结构

mkdir pictrack
cd pictrack
mkdir -p src/pictrack/
touch src/pictrack/__init__.py
mkdir tests
touch tests/__init__.py

% tree
.
├── README.md
├── src
│   └── pictrack
│       └── __init__.py
└── tests
    └── __init__.py

4 directories, 3 files

初始化项目虚拟环境：

poetry init

% cat pyproject.toml 
[tool.poetry]
name = "pictrack"
version = "0.1.0"
description = ""
authors = ["yongjiangshi <[email protected]>"]
readme = "README.md"

[tool.poetry.dependencies]
python = "^3.12"
fastapi = "^0.112.2"


[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"

编写 README.md 文件

安装开发包

poetry install

初始 Git 提交

git init 
git add . 
git commit -m "feat: First commit"

项目功能开发

创建命令行入口

安装依赖

poetry add click

创建cmdline.py

touch src/pictrack/cmdline.py

编辑 pyproject.toml ，将命令行入口注册到项目描述文件中：

[tool.poetry.scripts]
pictrack = "pictrack.cmdline:main"

提交代码

git add . 
git commit -m "feat: Add cmdline."

引入项目配置系统

poetry add dynaconf

建立配置包，和配置文件：

mkdir src/pictrack/config
touch src/pictrack/config/__init__.py
touch src/pictrack/config/settings.yml

提交代码:

git add .
git commit -m "feat: Add config."

引入日志

创建 src/pictrack/log.py ，初始化 log

from logging.config import dictConfig

from pictrack.config import settings


def init_log():
    log_config = {
        'version': 1,
        'disable_existing_loggers': False,
        'formatters': {
            'sample': {'format': '%(asctime)s %(levelname)s %(message)s'},
            'verbose': {'format': '%(asctime)s %(levelname)s %(name)s %(process)d %(thread)d %(message)s'},
            "access": {
                "()": "uvicorn.logging.AccessFormatter",
                "fmt": '%(asctime)s %(levelprefix)s %(client_addr)s - "%(request_line)s" %(status_code)s',
            },
        },
        'handlers': {
            "console": {
                "formatter": 'verbose',
                'level': 'DEBUG',
                "class": "logging.StreamHandler",
            },
        },
        'loggers': {
            '': {'level': settings.LOG_LEVEL, 'handlers': ['console']},
        },
    }

    dictConfig(log_config)
    ```

提交代码：
```bash
git add .
git commit -m "feat: Add log"

数据访问

安装依赖

poetry add sqlalchemy psycopg2-binary redis

激活虚拟环境

poetry shell

启动 Server

将本项目以可编辑方式安装到当前 Python 环境：

pip install -e .

命令行运行：

pictrack server

打包发布

打包

poetry build

发布

poetry publish

要使用Poetry将包发布到内部或私有PyPI服务器,可以按以下步骤操作:

  1. 配置自定义仓库

使用以下命令添加自定义仓库:

text

poetry config repositories.my_repository https://your-internal-pypi-server.com

poetry config repositories.pypi_dev http://127.0.0.1.8080/simple/

将 my_repository 替换为你想给这个仓库起的名字,URL替换为你的内部PyPI服务器地址。

  1. 配置认证信息

如果你的内部PyPI服务器需要认证,可以配置用户名和密码:

text

poetry config http-basic.my_repository username password

poetry config http-basic.pypi_dev pictrack pictrack

或者使用API令牌:

text

poetry config pypi-token.my_repository your-api-token

  1. 发布到自定义仓库

使用 -r 或 --repository 选项指定发布到哪个仓库:

text

poetry publish -r my_repository

poetry publish -r pypi_dev

如果你想在发布前构建,可以加上 --build 选项:

text

poetry publish --build -r my_repository

  1. 在pyproject.toml中配置

你也可以在 pyproject.toml 文件中配置发布仓库:

text

[tool.poetry.repositories] my_repository = { url = "https://your-internal-pypi-server.com" }

然后发布时就可以直接使用:

text

poetry publish -r my_repository

  1. 注意事项
  • 确保你有权限发布到指定的内部PyPI服务器
  • 发布前先测试包的构建和安装
  • 考虑使用版本控制和CI/CD流程来自动化发布过程

通过这些步骤,你就可以使用Poetry将包发布到指定的内部PyPI服务器了。这样可以确保你的私有包只在内部网络中可用,增强了安全性和控制性。

pip config set global.index-url http://server_ip:8080/simple 
pip config set global.trusted-host server_ip

正确publish

poetry config repositories.pypi_dev http://127.0.0.1:8080
poetry publish -r pypi_dev --verbose -u shi -p shi


 % poetry publish --build -r pypi_dev --verbose -u shi -p shi
There are 2 files ready for publishing. Build anyway? (yes/no) [no] y
Using virtualenv: /Users/macuser03/Library/Caches/pypoetry/virtualenvs/pictrack-xwTNEhgi-py3.12
Building pictrack (0.1.0)
  - Building sdist
  - Built pictrack-0.1.0.tar.gz
  - Building wheel
  - Built pictrack-0.1.0-py3-none-any.whl

Publishing pictrack (0.1.0) to pypi_dev
 - Uploading pictrack-0.1.0-py3-none-any.whl 100%
 - Uploading pictrack-0.1.0.tar.gz 100%