文本分段测试接口

开发环境

http://172.29.50.10:2333

POST

/zhixin_agent/knowledge_database/indexing_estimate

针对单文件，切分知识库里的文章，把大段文字拆成小段，做知识库配置部分的分段预览功能使用

请求参数

Body 参数application/json

示例

经济模式高质量模式

{
    "info_list": {
        "data_source_type": "upload_file",
        "file_info_list": {
            "file_ids": [
                "c969a6f5-89e6-41b6-8dbc-cff137ec53b5"
            ]
        }
    },
    "user_id": "admin",
    "process_rule": {
        "rules": {
            "pre_processing_rules": [
                {
                    "id": "remove_extra_spaces",
                    "enabled": true
                },
                {
                    "id": "remove_urls_emails",
                    "enabled": false
                }
            ],
            "segmentation": {
                "separator": "\n\n, \n, 。, !, ？, ",
                "max_tokens": 500,
                "chunk_overlap": 50
            }
        },
        "mode": "custom"
    },
    "indexing_technique": "economy",
    "doc_form": "text_model",
    "doc_language": "Chinese"
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

cURL

经济模式高质量模式

curl --location --request POST 'http://172.29.50.10:2333/zhixin_agent/knowledge_database/indexing_estimate' \
--header 'Content-Type: application/json' \
--data-raw '{
    "info_list": {
        "data_source_type": "upload_file",
        "file_info_list": {
            "file_ids": [
                "c969a6f5-89e6-41b6-8dbc-cff137ec53b5"
            ]
        }
    },
    "user_id": "admin",
    "process_rule": {
        "rules": {
            "pre_processing_rules": [
                {
                    "id": "remove_extra_spaces",
                    "enabled": true
                },
                {
                    "id": "remove_urls_emails",
                    "enabled": false
                }
            ],
            "segmentation": {
                "separator": "\n\n, \n, 。, !, ？, ",
                "max_tokens": 500,
                "chunk_overlap": 50
            }
        },
        "mode": "custom"
    },
    "indexing_technique": "economy",
    "doc_form": "text_model",
    "doc_language": "Chinese"
}'

返回响应

🟢200成功

application/json

Body

示例

{
    "total_segments": 0,
    "preview": [
        {
            "content": "string",
            "child_chunks": null
        }
    ],
    "qa_preview": null
}

修改于 2025-07-03 07:28:04

文件解析接口

二进制文件