# 安装所有格式支持
pip install 'markitdown[all]'

# 只装 PDF 支持
pip install 'markitdown[pdf]'

# 只装 Office 三件套
pip install 'markitdown[pptx,docx,xlsx]'

# PDF + DOCX + PPTX
pip install 'markitdown[pdf,docx,pptx]'

# 标准 Python
python -m venv .venv
source .venv/bin/activate  # Linux / macOS
.venv\Scripts\activate     # Windows

# 用 uv 创建（装包时用 uv pip install）
uv venv --python=3.12 .venv
source .venv/bin/activate

# 输出到 stdout
markitdown path-to-file.pdf > document.md

# 指定输出文件
markitdown path-to-file.pdf -o document.md

# 管道输入
cat path-to-file.pdf | markitdown

from markitdown import MarkItDown

md = MarkItDown()
result = md.convert("test.xlsx")
print(result.text_content)

# 查看已安装插件
markitdown --list-plugins

# 启用插件转换
markitdown --use-plugins path-to-file.pdf

# 查找社区插件：在 GitHub 搜 #markitdown-plugin

pip install markitdown-ocr
pip install openai  # 或任意 OpenAI 兼容客户端

from markitdown import MarkItDown
from openai import OpenAI

md = MarkItDown(
    enable_plugins=True,
    llm_client=OpenAI(),
    llm_model="gpt-4o",
)
result = md.convert("document_with_images.pdf")
print(result.text_content)

pip install markitdown-mcp
markitdown-mcp --help

markitdown path-to-file.pdf -o document.md -d -e "<document_intelligence_endpoint>"

from markitdown import MarkItDown

md = MarkItDown(docintel_endpoint="<your_endpoint>")
result = md.convert("test.pdf")
print(result.text_content)

pip install 'markitdown[az-content-understanding]'

markitdown path-to-file.pdf --use-cu --cu-endpoint "<content_understanding_endpoint>"

from markitdown import MarkItDown

# 零配置：按文件类型自动选择预置 analyzer
md = MarkItDown(cu_endpoint="<content_understanding_endpoint>")
result = md.convert("report.pdf")   # 文档 → prebuilt-document
result = md.convert("meeting.mp4")  # 视频 → prebuilt-video
result = md.convert("call.wav")     # 音频 → prebuilt-audio
print(result.markdown)

md = MarkItDown(
    cu_endpoint="<content_understanding_endpoint>",
    cu_analyzer_id="my-invoice-analyzer",
)
result = md.convert("invoice.pdf")
print(result.markdown)
# 输出带 YAML front matter 的提取字段：
# ---
# contentType: document
# fields:
#   VendorName: CONTOSO LTD.
#   InvoiceDate: '2019-11-15'
# ---

from markitdown.converters import ContentUnderstandingFileType

md = MarkItDown(
    cu_endpoint="<content_understanding_endpoint>",
    cu_file_types=[ContentUnderstandingFileType.PDF],  # 只有 PDF 走 CU
)

from markitdown import MarkItDown
from openai import OpenAI

client = OpenAI()

md = MarkItDown(
    llm_client=client,
    llm_model="gpt-4o",
    llm_prompt="用中文描述这张图片的内容",
)
result = md.convert("example.jpg")
print(result.text_content)

packages/
├── markitdown/               # 核心包
├── markitdown-mcp/           # MCP 服务器
├── markitdown-ocr/           # OCR 插件
└── markitdown-sample-plugin/ # 插件开发模板

输入文件（路径 / URL / 流）
    ↓
格式检测（扩展名 → MIME → 兜底分类）
    ↓
对应 Converter 处理
    ↓
Markdown 输出
    ↓
可选：LLM 增强（OCR / 图像描述）

from markitdown import MarkItDown

def extract_document(file_path: str) -> str:
    """RAG 管道的文档提取步骤"""
    md = MarkItDown(enable_plugins=True)
    result = md.convert(file_path)
    return result.text_content

content = extract_document("quarterly_report.pdf")

from markitdown import MarkItDown
from pathlib import Path

def batch_convert(directory: str, output_dir: str) -> None:
    """批量转换目录下所有支持的文件"""
    md = MarkItDown()
    output_path = Path(output_dir)
    output_path.mkdir(parents=True, exist_ok=True)

    for file in Path(directory).rglob("*"):
        if file.suffix.lower() in [".pdf", ".docx", ".pptx", ".xlsx"]:
            try:
                result = md.convert(str(file))
                (output_path / f"{file.stem}.md").write_text(result.text_content)
                print(f"ok: {file.name}")
            except Exception as e:
                print(f"fail: {file.name}: {e}")

batch_convert("./documents", "./markdown")

from markitdown import MarkItDown

md = MarkItDown()
with open("document.pdf", "rb") as f:
    result = md.convert_stream(f)
    print(result.text_content)

# 全量安装
pip install 'markitdown[all]'

# 常用组合
pip install 'markitdown[pdf,docx,pptx]'

# PDF + 图片 OCR
pip install 'markitdown[pdf]' markitdown-ocr openai

# Azure Document Intelligence
pip install 'markitdown[az-doc-intel]'

# Azure Content Understanding
pip install 'markitdown[az-content-understanding]'

# YouTube 视频转录
pip install 'markitdown[youtube-transcription]'