获取分段 - Dify Docs

GET

knowledge-fs

spaces

{knowledge_base_id}

documents

{document_id}

revisions

{revision}

chunks

{chunk_id}

获取分段

curl --request GET \
  --url https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id} \
  --header 'Authorization: Bearer <token>'

import requests

url = "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.get("https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}")
  .header("Authorization", "Bearer <token>")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

{
  "created_at": "2026-09-03T10:04:30Z",
  "document_id": "3d2f1c0b-9a8e-4d7c-b6a5-4f3e2d1c0b9a",
  "document_revision": 2,
  "enabled": true,
  "end_offset": 5120,
  "id": "7a6b5c4d-3e2f-4a1b-9c8d-7e6f5a4b3c2d",
  "kind": "chunk",
  "knowledge_space_id": "9b2e5b7a-3c1d-4e2f-8a9b-1c2d3e4f5a6b",
  "ordinal": 14,
  "parent_chunk_id": null,
  "parse_element_ids": [
    "el-40",
    "el-41"
  ],
  "section_path": [
    "3 Employment",
    "3.4 Termination"
  ],
  "start_offset": 4600,
  "text": "Either party may terminate this agreement with thirty days' written notice.",
  "token_count": 131,
  "user_metadata": {}
}

Authorizations

Authorization

string

header

required

每个请求都通过 API Key 认证：Authorization: Bearer {API_KEY}。应用接口使用应用 API Key，知识库接口使用知识库 API Key（快速开始）。

API Key 应保存在服务端，切勿嵌入客户端代码。缺失或无效的 Key 会返回 HTTP 401（unauthorized）。

Path Parameters

knowledge_base_id

string<uuid>

required

知识库 ID：即知识库 API 访问对话框中所示服务 API 端点的最后一段（…/knowledge-fs/spaces/<knowledge base ID>），也是地址栏中 /datasets/new/ 之后的那一段。

document_id

string<uuid>

required

文档 ID，从获取文档列表获取，或从已打开文档的地址栏中获取（/datasets/new/<knowledge base ID>/documents/<document ID>）。

revision

integer

required

修订版本号。从获取文档修订版本列表获取。

Required range: x >= 1

chunk_id

string<uuid>

required

分段 ID。从获取分段列表获取。

Response

分段。

偏移量基于解析后的文档文本，而非原始文件中的位置。

created_at

string

分段创建时间，ISO 8601 格式。

document_id

string

文档 ID，即获取文档中的 id。

document_revision

integer

分段所属的修订版本。每个修订版本有各自的分段，分段 ID 只能通过其所属的修订版本访问。

enabled

boolean

分段是否参与检索。此 API 中没有可修改该值的操作。

end_offset

integer | null

解析后文档文本中的 UTF-8 字节偏移量，从 0 开始计数，为结束位置（不包含）。未知时为 null。

string

分段 ID，用作获取分段的路径参数 chunk_id。

kind

string

分段的构建来源：chunk 表示一段正文（默认值），section 表示一个标题章节，table 表示表格，image 表示图片，summary 表示生成的摘要。

knowledge_space_id

string

知识库的内部 ID，并非路径中的知识库 ID。

ordinal

integer

分段在修订版本中的位置，从 0 开始。按此值排序即可按文档顺序读取分段。

parent_chunk_id

string | null

父子分段模式下父分段的 ID。顶层分段为 null。

parse_element_ids

string[]

构建该分段所用的解析元素 ID。

section_path

string[]

从文档根到该分段所在章节的各级标题。

start_offset

integer | null

解析后文档文本中的 UTF-8 字节偏移量，从 0 开始计数，为起始位置（包含）。未知时为 null。

text

string

分段内容。

token_count

integer

text 的 token 数。

user_metadata

object

附加在分段上的元数据，为自由格式的对象。

获取分段

curl --request GET \
  --url https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id} \
  --header 'Authorization: Bearer <token>'

import requests

url = "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.get("https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}")
  .header("Authorization", "Bearer <token>")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://{api_base_url}/knowledge-fs/spaces/{knowledge_base_id}/documents/{document_id}/revisions/{revision}/chunks/{chunk_id}")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body

{
  "created_at": "2026-09-03T10:04:30Z",
  "document_id": "3d2f1c0b-9a8e-4d7c-b6a5-4f3e2d1c0b9a",
  "document_revision": 2,
  "enabled": true,
  "end_offset": 5120,
  "id": "7a6b5c4d-3e2f-4a1b-9c8d-7e6f5a4b3c2d",
  "kind": "chunk",
  "knowledge_space_id": "9b2e5b7a-3c1d-4e2f-8a9b-1c2d3e4f5a6b",
  "ordinal": 14,
  "parent_chunk_id": null,
  "parse_element_ids": [
    "el-40",
    "el-41"
  ],
  "section_path": [
    "3 Employment",
    "3.4 Termination"
  ],
  "start_offset": 4600,
  "text": "Either party may terminate this agreement with thirty days' written notice.",
  "token_count": 131,
  "user_metadata": {}
}