输出格式与特性配置
output_formats
可传多个,指定需要返回的输出类型:
markdown— 结构化 Markdown 文本json— SoMark 兼容 JSON(pages[].blocks[])zip— 打包下载链接(含 md + json)
element_formats(JSON 字符串)
{
"image": "url", // url | base64 | file | none
"formula": "latex", // latex | mathml | ascii
"table": "html", // markdown | html | image
"cs": "image" // 化学结构式
}feature_config(JSON 字符串)
{
"enable_text_cross_page": false,
"enable_table_cross_page": false,
"enable_title_level_recognition": true,
"enable_inline_image": false,
"enable_table_image": true,
"enable_image_understanding": true,
"keep_header_footer": false
}- enable_title_level_recognition — 识别标题 H1/H2/H3 层级
- enable_table_cross_page — 跨页表格自动拼接
- enable_text_cross_page — 跨页段落自动拼接
- keep_header_footer — 是否保留页眉页脚块
JSON 输出结构
{
"pages": [{
"page_num": 0,
"page_size": { "w": 1190, "h": 1684 },
"blocks": [{
"idx": 0,
"type": "title",
"bbox": [72, 50, 540, 80],
"content": "第一章 引言",
"format": "text",
"title_level": 1
}]
}]
}