lanyuanxiaoyao
750ef50a8d
简化 parse_input() 为纯调度器，通过遍历 readers 的 supports() 方法识别输入类型，移除 URL 特殊处理和文件检查逻辑。各 reader 的 parse() 方法负责完整验证（文件存在、格式有效性）。 新增功能： - 添加 chardet 编码自动检测，支持多种中文编码回退机制 - 创建统一配置类管理编码、下载超时、日志等级等配置项 - HTML reader 支持本地文件编码检测和 URL 统一处理 安全性改进： - 修复 safe_open_zip() 路径遍历漏洞，使用 pathlib 规范化路径 - 添加边界检查，search_markdown() 检查负数参数 其他改进： - 修复类型注解（argparse.Namespace） - 日志系统仅输出 ERROR 级别，避免干扰 Markdown 输出 - 更新 BaseReader 接口文档，明确 supports() 和 parse() 职责划分 - 同步 delta specs 到主 specs（document-reading、html-reader、configuration、encoding-detection）
95 lines
1.7 KiB
TOML
95 lines
1.7 KiB
TOML
[project]
|
|
name = "lyxy-document"
|
|
version = "0.1.0"
|
|
description = "帮助AI工具读取转换文档到markdown的skill"
|
|
readme = "README.md"
|
|
requires-python = ">=3.11"
|
|
dependencies = [
|
|
"chardet>=5.0.0",
|
|
]
|
|
|
|
[project.optional-dependencies]
|
|
docx = [
|
|
"docling>=2.0.0",
|
|
"unstructured>=0.12.0",
|
|
"markitdown>=0.1.0",
|
|
"pypandoc-binary>=1.13.0",
|
|
"python-docx>=1.1.0",
|
|
"markdownify>=0.12.0",
|
|
]
|
|
xlsx = [
|
|
"docling>=2.0.0",
|
|
"unstructured>=0.12.0",
|
|
"markitdown>=0.1.0",
|
|
"pandas>=2.0.0",
|
|
"tabulate>=0.9.0",
|
|
]
|
|
pptx = [
|
|
"docling>=2.0.0",
|
|
"unstructured>=0.12.0",
|
|
"markitdown>=0.1.0",
|
|
"python-pptx>=0.6.0",
|
|
"markdownify>=0.12.0",
|
|
]
|
|
pdf = [
|
|
"docling>=2.0.0",
|
|
"unstructured>=0.12.0",
|
|
"unstructured-paddleocr>=0.1.0",
|
|
"markitdown>=0.1.0",
|
|
"pypdf>=4.0.0",
|
|
"markdownify>=0.12.0",
|
|
]
|
|
html = [
|
|
"trafilatura>=1.10.0",
|
|
"domscribe>=0.1.0",
|
|
"markitdown>=0.1.0",
|
|
"html2text>=2024.2.26",
|
|
"beautifulsoup4>=4.12.0",
|
|
]
|
|
http = [
|
|
"httpx>=0.27.0",
|
|
"pyppeteer>=2.0.0",
|
|
"selenium>=4.18.0",
|
|
]
|
|
office = [
|
|
"lyxy-document[docx,xlsx,pptx,pdf]",
|
|
]
|
|
web = [
|
|
"lyxy-document[html,http]",
|
|
]
|
|
full = [
|
|
"lyxy-document[office,web]",
|
|
]
|
|
dev = [
|
|
"pytest>=8.0.0",
|
|
"pytest-cov>=4.1.0",
|
|
"black>=24.0.0",
|
|
"isort>=5.13.0",
|
|
"mypy>=1.8.0",
|
|
]
|
|
|
|
[project.scripts]
|
|
lyxy-document-reader = "lyxy_document_reader:main"
|
|
|
|
[build-system]
|
|
requires = ["hatchling"]
|
|
build-backend = "hatchling.build"
|
|
|
|
[tool.black]
|
|
line-length = 100
|
|
target-version = ["py311"]
|
|
|
|
[tool.isort]
|
|
profile = "black"
|
|
line-length = 100
|
|
|
|
[tool.mypy]
|
|
python_version = "3.11"
|
|
warn_return_any = true
|
|
warn_unused_configs = true
|
|
disallow_untyped_defs = true
|
|
|
|
[tool.pytest.ini_options]
|
|
testpaths = ["tests"]
|
|
pythonpath = ["."]
|