### Start QAnything Services with Docker Compose

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

These commands navigate into the QAnything directory and start the project's services using Docker Compose, tailored for different operating systems. The startup process typically takes about 30 seconds until the backend service is ready.

```shell
cd QAnything
# Start on Linux
docker compose -f docker-compose-linux.yaml up
```

```shell
cd QAnything
# Start on Mac
docker compose -f docker-compose-mac.yaml up
```

```shell
cd QAnything
# Start on Windows
docker compose -f docker-compose-win.yaml up
```

--------------------------------

### Install Frontend Dependencies

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

These commands install all necessary project dependencies using either Yarn or npm. This is a crucial step before starting development or building the application.

```shell
yarn
# \u6216
npm install
```

--------------------------------

### Offline Docker Image and Code Setup for Windows

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

This sequence of commands facilitates offline deployment of QAnything on Windows. It involves downloading necessary Docker images on a networked machine, packaging them, downloading the QAnything source code, transferring both to an offline machine, loading the images, extracting the code, and finally starting the services.

```shell
# Download the docker image on a networked machine
docker pull quay.io/coreos/etcd:v3.5.5
docker pull minio/minio:RELEASE.2023-03-20T20-16-18Z
docker pull milvusdb/milvus:v2.4.8
docker pull mysql:8.4
docker pull xixihahaliu01/qanything-win:v1.5.1
```

```shell
# pack image
docker save quay.io/coreos/etcd:v3.5.5 minio/minio:RELEASE.2023-03-20T20-16-18Z milvusdb/milvus:v2.4.8 mysql:8.4 xixihahaliu01/qanything-win:v1.5.1 -o qanything_offline.tar
```

```shell
# download QAnything code
wget https://github.com/netease-youdao/QAnything/archive/refs/heads/master.zip
```

```shell
# Copy the image qanything_offline.tar and the code qany-master.zip to the offline machine 
cp QAnything-master.zip qanything_offline.tar /path/to/your/offline/machine
```

```shell
# Load image on offline machine 
docker load -i qanything_offline.tar
```

```shell
# Unzip the code and run it
unzip QAnything-master.zip
cd QAnything-master
docker compose -f docker-compose-win.yaml up
```

--------------------------------

### Clone QAnything Git Repository

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

This command clones the QAnything project repository from GitHub to your local machine, providing access to the source code and Docker configurations necessary for setup.

```shell
git clone https://github.com/netease-youdao/QAnything.git
```

--------------------------------

### Serve Built Frontend Application

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

These commands start a local static server to host the previously built QAnything frontend application from the 'dist/qanything' directory. This is useful for testing the production build locally.

```shell
yarn serve
# \u6216
npm run serve
```

--------------------------------

### Start Development Web Server

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

These commands launch the QAnything frontend application in development mode, typically with hot-reloading and debugging features enabled. The service will be accessible locally.

```shell
yarn dev
# \u6216
npm run dev
```

--------------------------------

### Get All Knowledge Base Status Request Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

A Python example demonstrating how to query the status of all knowledge bases for a given user ID via a POST request.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/get_total_status"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp"
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### QAnything API Access Details

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

Provides the base URL for accessing the QAnything API. Detailed API documentation is available in a separate markdown file, which should be consulted for specific endpoints, parameters, and responses.

```APIDOC
API address: http://localhost:8777/qanything/
For detailed API documentation, please refer to [QAnything API documentation](docs/API.md)
```

--------------------------------

### QAnything Debug Log File Descriptions

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

A comprehensive list of log files generated by the QAnything project, detailing the purpose of each log to assist with debugging and monitoring various service components.

```APIDOC
debug.log: User request processing log
```

```APIDOC
sanic_api.log: Backend service running log
```

```APIDOC
llm_embed_rerank_tritonserver.log (Single card deployment): LLM embedding and rerank tritonserver service startup log
```

```APIDOC
llm_tritonserver.log (Multi-card deployment): LLM tritonserver service startup log
```

```APIDOC
embed_rerank_tritonserver.log (Multi-card deployment or use of the OpenAI interface.): Embedding and rerank tritonserver service startup log
```

```APIDOC
rerank_server.log: Rerank service running log
```

```APIDOC
ocr_server.log: OCR service running log
```

```APIDOC
npm_server.log: Front-end service running log
```

```APIDOC
llm_server_entrypoint.log: LLM intermediate server running log
```

```APIDOC
fastchat_logs/*.log: FastChat service running log
```

--------------------------------

### Close QAnything Docker Services

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/README.md

Instructions for gracefully shutting down QAnything services. For services started in the foreground (without -d), simply press Ctrl+C. For services running in detached mode, use the 'docker compose down' command to stop and remove containers.

```shell
# Front desk service startup mode like:
docker compose -f docker-compose-xxx.yaml up  # To close the service, please press Ctrl+C.
```

```shell
# Backend service startup mode like: 
docker compose -f docker-compose-xxx.yaml up -d # To close the service, please execute the following command.
docker compose -f docker-compose-xxx.yaml down
```

--------------------------------

### Check Node.js Version

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

This command displays the currently installed version of Node.js on your system. It is recommended to use Node.js version 18.16.0 for QAnything frontend development.

```shell
node -v
```

--------------------------------

### Configure npm Registry to Taobao Mirror

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

This command changes the default npm package registry to the Taobao mirror, which can significantly speed up dependency downloads in certain regions. This is an optional step if 'npm install' is slow.

```shell
npm config set registry https://registry.npmmirror.com
```

--------------------------------

### Build QAnything Frontend for Production

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/front_end/README.md

These commands compile and bundle the QAnything frontend application for production deployment. The output is typically placed in a 'dist/qanything' directory, ready for static serving.

```shell
yarn build
# \u6216
npm run build
```

--------------------------------

### Clean Knowledge Base Request Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Provides a Python code example for initiating a cleanup operation on knowledge base files, targeting files by user ID and their processing status (e.g., 'gray').

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/clean_files_by_status"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp",
    "status": "gray"
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Get All Knowledge Base Status Response Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Presents the JSON response structure for the knowledge base status query, detailing the status of individual knowledge bases (green, yellow, red, gray counts).

```json
{
  "code": 200,
  "status": {
    "zzp": {
      "默认知识库KB0015df77a8eb46f6951de513392dc250": {
        // kb_name + kb_id
        "green": 10,
        "yellow": 0,
        "red": 0,
        "gray": 0
      },
      "知识问答KB6a4534f753b54c4198b62770e26b1a92": {
        "green": 16,
        "yellow": 0,
        "red": 0,
        "gray": 0
      }
    }
  }
}
```

--------------------------------

### Python Example: Upload FAQs

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Demonstrates how to send a POST request to upload FAQs using the Python `requests` library. It includes the user ID, knowledge base ID, and a list of FAQs in the request body.

```python
import os
import requests

url = "http://{your_host}:8777/api/local_doc_qa/upload_faqs"
data = {
    "user_id": "zzp",
    "kb_id": "KB6dae785cdd5d47a997e890521acbe1c9_FAQ",
    "faqs": [{"question": "北京3月13日天气", "answer": "15-29度，小雨转晴"}]
}

response = requests.post(url, data=data)
print(response.text)
```

--------------------------------

### Delete Knowledge Base Request Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Provides a Python example for making a POST request to delete knowledge bases, specifying user ID and a list of knowledge base IDs.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/delete_knowledge_base"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp", //用户id
    "kb_ids": [
        "KB1cd81f2bc515437294bda1934a20b235"
    ] //知识库id列表
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Get File List API Request Example (Python)

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Python code demonstrating how to make a POST request to the /list_files endpoint. It sends the user_id and kb_id in the request body to retrieve a list of files associated with a specific knowledge base.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/list_files"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp",  //用户id
    "kb_id": "KBb1dd58e8485443ce81166d24f6febda7" //知识库id
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Upload Web File API Response Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Example JSON response for a successful web file upload operation, showing file details like ID, name, URL, status, and timestamp. Note that bytes are 0 for web files.

```json
{
  "code": 200,
  "msg": "success，后台正在飞速上传文件，请耐心等待",
  "data": [
    {
      "file_id": "9a49392e633d4c6f87e0af51e8c80a86",
      "file_name": "index.html.web",
      "file_url": "https://ai.youdao.com/DOCSIRMA/html/trans/api/wbfy/index.html",
      "status": "gray",
      "bytes": 0,
      "timestamp": "202401261809"
    }
  ]
}
```

--------------------------------

### Retrieve File Base64 API Response Example (JSON)

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

This JSON snippet provides an example of a successful response from the 'get_file_base64' API. It includes a status code, a success message, and the 'file_base64' field containing the base64 encoded file content.

```json
{
    "code": 200,
    "msg": "success",
    "file_base64": "xxx"
}
```

--------------------------------

### Clean Knowledge Base Response Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Illustrates the JSON response received after a successful knowledge base cleanup operation, including a success message and a list of cleaned file names.

```json
{
  "code": 200,
  "msg": "delete gray files success",
  "data": [
    "升学百科benchmark_20231120.xlsx_20231121235103.txt",
    "网页翻译-API文档.md",
    "api_inter_virtualPerson_execution_3451_executionrersult_4699_10_stats.csv",
    "中文成语-如火如荼.png",
    "有道知识库问答产品介绍.pptx",
    "韦小宝身份证.jpg",
    "fake-email.eml",
    "有道领世辅导日常问题处理手册.pdf",
    "xx案件支付三者车损人伤保险赔款及权益转让授权书.docx"
  ]
}
```

--------------------------------

### Delete File Request Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Demonstrates how to send a POST request to delete a specific file from a knowledge base using the QAnything API.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/delete_files"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp", //用户id
    "kb_id": "KB1271e71c36ec4028a6542586946a3906", //知识库id
    "file_ids": [
        "73ff7cf76ff34c8aa3a5a0b4ba3cf534"
    ] //文件id列表
}
response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Python Example for Streaming Q&A Requests

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

This Python script demonstrates how to send a streaming question-answering request to the QAnything API. It uses the 'requests' library to make a POST request to the '/api/local_doc_qa/local_doc_chat' endpoint and processes the streamed response line by line, decoding JSON chunks.

```python
import os
import json
import requests
import time
import random
import string
import argparse

def stream_requests(data_raw):
    url = 'http://{your_host}:8777/api/local_doc_qa/local_doc_chat'
    response = requests.post(
        url,
        json=data_raw,
        timeout=60,
        stream=True
    )
    for line in response.iter_lines(decode_unicode=False, delimiter=b"\n\n"):
        if line:
            yield line

def test():
    data_raw = {
      "kb_ids": ["KB633c69d07a2446b39b3a38e3628b8ada"],
      "question": "你好",
      "user_id": "zzp",
      "streaming": True,
      "history": []
    }
    for i, chunk in enumerate(stream_requests(data_raw)):
        if chunk:
            chunkstr = chunk.decode("utf-8")[6:]
            chunkjs = json.loads(chunkstr)
            print(chunkjs)

if __name__ == '__main__':
    test()
```

--------------------------------

### List Knowledge Base API Request Example (Python)

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Python code demonstrating how to make a POST request to the /list_knowledge_base endpoint. It sets the Content-Type header and sends the user_id in the request body to retrieve available knowledge bases.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/list_knowledge_base"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp"
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Delete File Response Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Illustrates the expected JSON response format after successfully deleting a file, including status code and a success message.

```json
{
  "code": 200, //状态码
  "msg": "documents ['73ff7cf76ff34c8aa3a5a0b4ba3cf534'] delete success" //提示信息
}
```

--------------------------------

### Delete Knowledge Base Response Example

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Shows the JSON response format for a successful knowledge base deletion operation, including status and a confirmation message.

```json
{
  "code": 200, //状态码
  "msg": "Knowledge Base [('KB1cd81f2bc515437294bda1934a20b235',)] delete success" //提示信息
}
```

--------------------------------

### Python Example: Update Bot Information

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Demonstrates how to send a POST request to update bot information using the Python `requests` library. It includes setting the content type header and constructing the JSON payload with various bot attributes, including `kb_ids` as a list.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/update_bot"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp",
    "bot_id": "BOTc87cc749a9844325a2e26f09bf8f6918",
    "bot_name": "测试Bot2",
    "description": "测试3",
    "head_image": "xx",
    "prompt_setting": "测试4",
    "welcome_message": "测试5",
    "model": "gpt",
    "kb_ids": ["KB1", "KB2"]
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Modify Ollama Model Context Length for Improved QAnything Performance

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/FAQ_zh.md

This snippet provides a step-by-step guide to manually increase the context window size for an Ollama model, such as 'qwen2:72b-instruct'. This is necessary because Ollama's default context length (e.g., 2048) can truncate relevant information, negatively impacting QAnything's question-answering performance. The process involves pulling the model, exporting its Modelfile, editing the 'num_ctx' parameter, and creating a new model with the updated configuration.

```bash
ollama pull qwen2:72b-instruct

ollama show --modelfile qwen2:72b-instruct > Modelfile

vim Modelfile  // 加一行：PARAMETER num_ctx 32000

ollama create -f Modelfile qwen2:72b_ctx32k
```

--------------------------------

### Python Example: Delete Bot

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Illustrates how to send a POST request to delete a bot using the Python `requests` library. It sets the content type header and includes the user ID and bot ID in the JSON payload.

```python
import requests
import json

url = "http://{your_host}:8777/api/local_doc_qa/delete_bot"
headers = {
    "Content-Type": "application/json"
}
data = {
    "user_id": "zzp",
    "bot_id": "BOTc87cc749a9844325a2e26f09bf8f6918"
}

response = requests.post(url, headers=headers, data=json.dumps(data))

print(response.status_code)
print(response.text)
```

--------------------------------

### Example of Streaming Q&A API Response

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

This text snippet illustrates the typical format of a streaming response from the QAnything API. Each line represents a JSON chunk, often containing incremental parts of the answer or status updates, culminating in the final complete response object.

```text
{'code': 200, 'msg': 'success', 'question': '', 'response': '', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '你', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '好', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '！', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '我', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '是', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '有', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '道', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '开', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '发', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '的', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '大', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '模', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '型', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '。', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '有', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '什', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '么', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '问题', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '我', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '可以', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '帮', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '助', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '你', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '解', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '答', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '吗', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '？', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '', 'response': '', 'history': [], 'source_documents': []}
{'code': 200, 'msg': 'success', 'question': '你好', 'response': '你好！我是有道开发的大模型。有什么问题我可以帮助你解答吗？', 'history': [['你好', '你好！我是有道开发的大模型。有什么问题我可以帮助你解答吗？']], 'source_documents':[]}
```

--------------------------------

### Get All Knowledge Base Status API Endpoint

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

Describes the API endpoint for retrieving the status of all knowledge bases associated with a user, including the URL and required parameters.

```APIDOC
Endpoint: POST /api/local_doc_qa/get_total_status
URL: http://{your_host}:8777/api/local_doc_qa/get_total_status
Body Parameters:
  user_id (String, Required): 用户 id (Example: "zzp")
```

--------------------------------

### Get File List API Reference

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

API documentation for retrieving a list of files within a specified knowledge base. It details the endpoint, various request parameters for filtering and pagination, and the comprehensive structure of the JSON response, including file statuses and details.

```APIDOC
URL: http://{your_host}:8777/api/local_doc_qa/list_files
Method: POST

Request Body Parameters:
  user_id: String (required) - 用户 id (User ID)
  kb_id: String (required) - 知识库 id (Knowledge base ID)
  file_id: String (optional) - 文件id，指定文件id则只返回单个文件的状态 (File ID, returns status of single file if specified)
  page_offset: Integer (optional, default: 1) - 默认值为1，结果太多时可指定分页id（从1开始），与page_limit配合使用 (Default 1, pagination offset)
  page_limit: Integer (optional, default: 10) - 默认值为10，结果太多时可指定分页id（从1开始），与page_offset配合使用 (Default 10, pagination limit)

Response Body Schema:
  code: Integer - 状态码 (Status code)
  msg: String - 提示信息 (Message)
  data: Object
    status_count: Object - 知识库所有文件状态 (All file statuses in knowledge base)
      green: Integer
      red: Integer
      gray: Integer
      yellow: Integer
    details: Array of FileDetails - 每个文件的具体状态 (Specific status for each file)
      FileDetails:
        file_id: String - 文件id (File ID)
        file_name: String - 文件名 (File name)
        status: String - 文件状态（red：入库失败-切分失败，green，成功入库，yellow：入库失败-milvus失败，gray：正在入库） (File status: red-failed segmentation, green-success, yellow-failed milvus, gray-processing)
        bytes: Integer - File size in bytes
        chunks_number: Integer - Number of chunks
        content_length: Integer - 文件解析后字符长度，用len()计算 (Character length after file parsing)
        file_location: String - File storage path
        timestamp: String - Timestamp
        msg: String - Status message
    page_id: Integer - Current page ID
    total: Integer - Total number of files
    total_page: Integer - Total number of pages
```

--------------------------------

### Get File Base64 API Endpoint Documentation

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

This section documents the API endpoint for retrieving a file's content in base64 format. It specifies the URL, required request parameters (file_id), their types, and descriptions.

```APIDOC
URL: http://{your_host}:8777/api/local_doc_qa/get_file_base64

Request Parameters (Body):
- Parameter Name: file_id
  - Example Value: "698428590b1e45108b63b943fac16559"
  - Required: Yes
  - Type: String
  - Description: File ID
```

--------------------------------

### Retrieve Document Full Content Response (JSON)

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

This JSON example illustrates the successful response structure from the `/api/local_doc_qa/get_doc_completed` endpoint. It includes the HTTP status code, a success message, the `completed_text` of the document, and a list of `chunks` with their content and metadata.

```json
{
    "code": 200,
    "msg": "success",
    "completed_text": "\n6.全国两会会议精神.docx\n全国政协十四届二次会议开幕会会议精神\n\n6.全国两会会议精神.docx\n中国人民政治协商会议第十四届全国委员会第二次会议4日下午在人民大会堂开幕。2000多名全国政协委员将紧紧围绕中共中央决策部署，牢记政治责任，积极建言资政，广泛凝聚共识，共同谱写中国式现代化的壮美华章.\n\n6.全国两会会议精神.docx\n大会首先审议通过了政协第十四届全国委员会第二次会议议程.\n\n6.全国两会会议精神.docx\n王沪宁代表政协第十四届全国委员会常务委员会，向大会报告工作.\n\n6.全国两会会议精神.docx\n王沪宁表示，2023年是全面贯彻落实中共二十大精神的开局之年。以习近平同志为核心的中共中央团结带领全党全国各族人民，坚持稳中求进工作总基调，接续奋斗、砥砺前行，坚决克服内外困难，全面深化改革开放，新冠疫情防控平稳转段，高质量发展扎实推进，科技创新实现新突破，安全发展基础巩固夯实，民生保障有力有效，经济社会发展主要预期目标圆满完成，社会大局保持稳定，全面建设社会主义现代化国家迈出坚实步伐，极大增强了全国各族人民信心和底气.\n\n6.全国两会会议精神.docx\n王沪宁总结了过去一年来人民政协工作。他说，在以习近平同志为核心的中共中央坚强领导下，政协全国委员会及其常务委员会坚持以习近平新时代中国特色社会主义思想为指导，全面贯彻中共二十大和二十届二中全会精神，深刻领悟“两个确立”的决定性意义，增强“四个意识”、坚定“四个自信”、做到“两个维护”，坚持团结和民主两大主题，坚持人民政协性质定位，坚持在党和国家工作大局中谋划推进政协工作。坚持中国共产党对人民政协工作的全面领导、把牢履职正确政治方向，强化政治培训、提高政治能力和履职本领，完善工作制度体系、夯实履职制度基础，贯彻中共中央大兴调查研究要求、提高调研议政质量，围绕中共二十大重大部署协商议政、服务党和国家中心任务，践行以人民为中心的发展思想、助推保障和改善民生，守正创新、团结奋进，在历届全国政协打下的良好基础上，各项工作取得新成效，服务党和国家事业发展作出新贡献.\n\n6.全国两会会议精神.docx\n王沪宁表示，2024年是中华人民共和国成立75周年，是实现“十四五”规划目标任务的关键一年，也是人民政协成立75周年。人民政协要坚持以习近平新时代中国特色社会主义思想为指导，全面贯彻中共二十大和二十届二中全会精神，坚持党的领导、统一战线、协商民主有机结合，紧紧围绕推进中国式现代化履职尽责，推进思想政治引领，积极建言资政，广泛凝聚共识，加强自身建设，为实现全年经济社会发展目标任务汇聚智慧和力量....\n",
    "chunks": [
        {
            "page_content": "6.全国两会会议精神.docx\n全国政协十四届二次会议开幕会会议精神",
            "metadata": {
                "user_id": "zzp",
                "kb_id": "KB70b80588e3614fb9ac21581d38176a
```

--------------------------------

### JSON Response: Delete Bot Success

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

An example of the successful JSON response returned after a bot deletion operation. It indicates a 200 status code and a success message.

```json
{
    "code": 200,
    "msg": "Bot BOTc87cc749a9844325a2e26f09bf8f6918 delete success"
}
```

--------------------------------

### JSON Response: Update Bot Success

Source: https://github.com/netease-youdao/qanything/blob/qanything-v2/docs/API.md

An example of the successful JSON response returned after a bot update operation. It indicates a 200 status code and a success message.

```json
{
    "code": 200,
    "msg": "Bot BOTc87cc749a9844325a2e26f09bf8f6918 update success"
}
```