Emoji Mode Integration Guide

Version Information

• Document Version: v1.0
• Updated Date: 2024-10-26
• Applicable Service Version: tcpv2.3+

---

Overview

Emoji Mode is a brand new feature that allows the server to automatically match and send corresponding emoji identifiers based on conversation content. Clients can display custom emoji animations, images, or other visual effects based on these identifiers.

New Additions

• ✨ New Message Type: EMOJI (Type Value = 9)
• 🎨 Multiple Matching Modes: false, true, dimi, dimi2, jfb
• 🚀 Smart Matching: Supports both sentiment analysis and keyword matching algorithms

---

Authentication Parameter Configuration

emoji_mode Parameter

Add the emoji_mode parameter in the TCP authentication message:

##START[AUTH][task_id][seq]your_jwt_token##emoji_mode:true##END

Or in HTTP API (if you are using HTTP interface):

{
  "emoji_mode": "true"
}

Parameter Description

Parameter Value	Description	Prompt Match	Answer Match	Matching Algorithm	Prompt with playSound
false or not passed	Disable emoji function	❌	❌	-	-
true	Answer-only matching	❌	✅	Sentiment Analysis	-
dimi	Match both question and answer	✅	✅	Keyword Matching	✅ playSound:1
dimi2	Match both question and answer	✅	✅	Keyword Matching	❌
jfb	Match both question and answer (5 emotion categories)	✅	✅	JFB Sentiment Analysis (angry/sad/hesitant/anxious/excited)	❌

Compatibility Notes

• ✅ Backward Compatible: Older client versions default to false if this parameter is not passed
• ✅ Flexible Switching: Mode can be dynamically switched during each authentication
• ✅ Boolean Compatibility: Both true/false and "true"/"false" are supported

---

Message Format

Message Type Definition

MessageType.EMOJI = 9

Protocol Format

##START + Header + JSON_Data + ##END

Header Structure (13 bytes)

Field	Length	Description
msg_type	1 byte	Message Type = 9
task_id	8 bytes	Task ID (UTF-8 encoded)
seq	4 bytes	Sequence Number (fixed as "0000")

JSON_Data Structure

{
  "emoji": "emoji_key"
}

Complete Example

Assuming task_id = abc12345, the matched emoji is happy:

##START[0x09][abc12345][0000]{"emoji":"happy"}##END

Binary representation:

0x23 0x23 0x53 0x54 0x41 0x52 0x54 // ##START (7 bytes)
0x09                                           // msg_type = 9 (1 byte)
0x61 0x62 0x63 0x31 0x32 0x33 0x34 0x35       // task_id = "abc12345" (8 bytes)
0x30 0x30 0x30 0x30                            // seq = "0000" (4 bytes)
{"emoji":"happy"}                              // JSON data (variable length)
0x23 0x23 0x45 0x4E 0x44                       // ##END (5 bytes)

---

Emoji Identifier Reference

True Mode - Sentiment-based Emojis (20 types)

Based on sentiment analysis algorithm, returns the following 20 emotion identifiers:

Emoji Key	Description	Trigger Examples
happy	Happy	"Great!" / "Awesome"
laughing	Laughing	"HAHAHA" / "So funny"
sad	Sad	"Feeling down" / "Disappointed"
angry	Angry	"So mad" / "Hate it"
crying	Crying	"Wanna cry" / "Sobbing"
loving	Loving	"Love you" / "Like it"
surprised	Surprised	"Oh my" / "Wow"
shocked	Shocked	"Can't believe" / "Scared"
thinking	Thinking	"?" / "Let me think"
embarrassed	Embarrassed	"Sorry" / "Shy"
winking	Winking	"You know what I mean" / "Naughty"
cool	Cool	"Impressive" / "Awesome"
relaxed	Relaxed	"Comfortable" / "Pleasant"
delicious	Delicious	"Yummy" / "Tasty"
kissy	Kiss	"Mua" / "Love kiss"
confident	Confident	"Of course" / "Definitely"
sleepy	Sleepy	"Good night" / "So tired"
silly	Silly	"Silly goose" / "Dopey"
confused	Confused	"Don't understand" / "Puzzled"
funny	Funny	"Humorous" / "Hilarious"

Dimi Mode - Keyword-based Emojis (2868+ types)

Based on keyword matching, supports a large number of scenario-based emojis, for example:

Emoji Key	Keyword Examples
hu_die	"go for a walk", "spring walk"
sheng_bing	"sore throat", "喉咙痛"
xia_yu	"after the rain", "下雨"
bi_xin	"比心", "heart gesture"
kai_hua	"peach blossoms", "开花"
...	...

💡 Tip: Dimi mode supports mixed Chinese/English matching, prioritizing longer keywords.

JFB Mode - 5 Emotion Category Emojis

Data file: emoji_service/JFB_emoji_mode.json, 5 categories total, no emoji sent if no match (returns None):

Emoji Key	Description	Trigger Examples
angry	Angry	"So mad" / "Hate it"
sad	Sad	"Sad" / "Wanna cry"
hesitant	Hesitant	"Don't know what to do" / "?" / "..."
anxious	Anxious	"Worried" / "Nervous"
excited	Excited	"Great!" / "Awesome"

💡 Tip: JFB mode punctuation rules — ? alone → hesitant; ... → hesitant; ! + positive content → excited; ! + negative content → angry.

---

Client Implementation Examples

1. Parse Emoji Message

Python Example

import struct
import json

START_MARK = b'##START'
END_MARK = b'##END'

def parse_emoji_message(data):
    """Parse EMOJI message"""
    if data.startswith(START_MARK) and data.endswith(END_MARK):
        # Remove markers
        frame = data[len(START_MARK):-len(END_MARK)]
        
        # Parse header
        msg_type = frame[0]
        task_id = frame[1:9].decode('utf-8')
        seq = frame[9:13].decode('utf-8')
        
        # Check if it's an EMOJI message
        if msg_type == 9:
            # Parse JSON data
            json_data = frame[13:].decode('utf-8')
            emoji_info = json.loads(json_data)
            return {
                'type': 'emoji',
                'task_id': task_id,
                'emoji_key': emoji_info['emoji']
            }
    
    return None

# Usage Example
data = b'##START\x09abc123450000{"emoji":"happy"}##END'
result = parse_emoji_message(data)
print(result)
# Output: {'type': 'emoji', 'task_id': 'abc12345', 'emoji_key': 'happy'}

Unity C# Example

using System;
using System.Text;
using Newtonsoft.Json.Linq;

public class EmojiMessage
{
    public string Type { get; set; }
    public string TaskId { get; set; }
    public string EmojiKey { get; set; }
}

public class MessageParser
{
    private static readonly byte[] START_MARK = Encoding.UTF8.GetBytes("##START");
    private static readonly byte[] END_MARK = Encoding.UTF8.GetBytes("##END");
    
    public static EmojiMessage ParseEmojiMessage(byte[] data)
    {
        // Check start and end markers
        if (!StartsWith(data, START_MARK) || !EndsWith(data, END_MARK))
            return null;
        
        // Extract frame data
        int frameStart = START_MARK.Length;
        int frameEnd = data.Length - END_MARK.Length;
        
        // Parse message type
        byte msgType = data[frameStart];
        if (msgType != 9)
            return null; // Not an EMOJI message
        
        // Parse task_id (8 bytes)
        string taskId = Encoding.UTF8.GetString(data, frameStart + 1, 8);
        
        // Parse JSON data
        int jsonStart = frameStart + 13; // 1 + 8 + 4
        int jsonLength = frameEnd - jsonStart;
        string jsonData = Encoding.UTF8.GetString(data, jsonStart, jsonLength);
        
        // Parse JSON
        JObject json = JObject.Parse(jsonData);
        
        return new EmojiMessage
        {
            Type = "emoji",
            TaskId = taskId,
            EmojiKey = json["emoji"].ToString()
        };
    }
    
    private static bool StartsWith(byte[] data, byte[] prefix)
    {
        if (data.Length < prefix.Length) return false;
        for (int i = 0; i < prefix.Length; i++)
            if (data[i] != prefix[i]) return false;
        return true;
    }
    
    private static bool EndsWith(byte[] data, byte[] suffix)
    {
        if (data.Length < suffix.Length) return false;
        int offset = data.Length - suffix.Length;
        for (int i = 0; i < suffix.Length; i++)
            if (data[offset + i] != suffix[i]) return false;
        return true;
    }
}

2. Display Emoji Animation

Example: Show Corresponding Animation Based on emoji_key

public class EmojiDisplay : MonoBehaviour
{
    [SerializeField] private Dictionary<string, AnimationClip> emojiAnimations;
    [SerializeField] private Animator characterAnimator;
    
    public void ShowEmoji(string emojiKey)
    {
        if (emojiAnimations.ContainsKey(emojiKey))
        {
            // Play corresponding expression animation
            characterAnimator.Play(emojiKey);
            
            // Or show emoji icon
            ShowEmojiIcon(emojiKey);
        }
        else
        {
            Debug.LogWarning($"Emoji not found: {emojiKey}");
        }
    }
    
    private void ShowEmojiIcon(string emojiKey)
    {
        // Display expression bubble above character
        // Implement your own display logic here
    }
}

3. Complete Receive Flow

import socket
import threading

class TCPClient:
    def __init__(self, host, port):
        self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        self.sock.connect((host, port))
        self.buffer = b""
    
    def receive_loop(self):
        """Message receive loop"""
        while True:
            data = self.sock.recv(1024)
            if not data:
                break
            
            self.buffer += data
            
            # Process complete messages in buffer
            while b'##END' in self.buffer:
                frame, self.buffer = self.buffer.split(b'##END', 1)
                
                if b'##START' in frame:
                    frame = frame[frame.index(b'##START'):]
                    frame += b'##END'  # Add back the delimiter
                    
                    # Parse message
                    self.handle_message(frame)
    
    def handle_message(self, frame):
        """Handle single message"""
        if len(frame) < 20:
            return
        
        msg_type = frame[7]  # Skip ##START
        
        if msg_type == 4:  # TEXT
            # Handle text message
            self.handle_text(frame)
        elif msg_type == 2:  # AUDIO_FRAME
            # Handle audio message
            self.handle_audio(frame)
        elif msg_type == 9:  # EMOJI
            # Handle emoji message
            emoji_msg = parse_emoji_message(frame)
            if emoji_msg:
                print(f"Received emoji: {emoji_msg['emoji_key']}")
                self.show_emoji(emoji_msg['emoji_key'])
    
    def show_emoji(self, emoji_key):
        """Show emoji (implement by yourself)"""
        pass

---

Message Timing

True Mode Flow

Client                                              Server
  |                                                    |
  |----(1) Send AUTH------------------------------>|
  |      emoji_mode: true                             |
  |                                                    |
  |<---(2) Send AUTH Success-------------------------|
  |                                                    |
  |----(3) Send User Input-------------------------->|
  |      "Hello!"                                     |
  |                                                    |
  |<---(4) TEXT: "Hello!"----------------------------|
  |<---(5) EMOJI: "happy"---------------------------|  ← Answer-only matching
  |<---(6) TEXT: "Nice to meet you"-----------------|
  |<---(7) EMOJI: "happy"---------------------------|
  |<---(8) END_FRAME--------------------------------|

Dimi Mode Flow

Client                                              Server
  |                                                    |
  |----(1) Send AUTH------------------------------>|
  |      emoji_mode: dimi                            |
  |                                                    |
  |<---(2) Send AUTH Success-------------------------|
  |                                                    |
  |----(3) Send User Input-------------------------->|
  |      "It's raining today"                         |
  |                                                    |
  |<---(4) EMOJI: "xia_yu"---------------------------|  ← Matches question
  |<---(5) TEXT: "Remember to bring an umbrella"---|
  |<---(6) EMOJI: "care"---------------------------|  ← Matches answer
  |<---(7) END_FRAME--------------------------------|

true_emoji_mode

{
    "emotion_keywords": {
        "happy": [
            "开心", "高兴", "快乐", "愉快", "幸福", "满意", "棒", "好", "不错",
            "完美", "棒极了", "太好了", "好呀", "好的",
            "happy", "joy", "great", "good", "nice", "awesome", "fantastic", "wonderful"
        ],
        "laughing": [
            "哈哈", "哈哈哈", "呵呵", "嘿嘿", "嘻嘻", "笑死", "太好笑了", "笑死我了",
            "lol", "lmao", "haha", "hahaha", "hehe", "rofl", "funny", "laugh"
        ],
        "funny": [
            "搞笑", "滑稽", "逗", "幽默", "笑点", "段子", "笑话", "太逗了",
            "hilarious", "joke", "comedy"
        ],
        "sad": [
            "伤心", "难过", "悲哀", "悲伤", "忧郁", "郁闷", "沮丧", "失望",
            "想哭", "难受", "不开心", "唉", "呜呜",
            "sad", "upset", "unhappy", "depressed", "sorrow", "gloomy"
        ],
        "angry": [
            "生气", "愤怒", "气死", "讨厌", "烦人", "可恶", "烦死了", "恼火",
            "暴躁", "火大", "气炸了",
            "angry", "mad", "annoyed", "furious", "pissed", "hate"
        ],
        "crying": [
            "哭泣", "泪流", "大哭", "伤心欲绝", "泪目", "流泪", "哭死", "哭晕",
            "想哭", "泪崩",
            "cry", "crying", "tears", "sob", "weep"
        ],
        "loving": [
            "爱你", "喜欢", "爱", "亲爱的", "宝贝", "么么哒", "抱抱", "想你",
            "思念", "最爱", "亲亲", "喜欢你",
            "love", "like", "adore", "darling", "sweetie", "honey", "miss you", "heart"
        ],
        "embarrassed": [
            "尴尬", "不好意思", "害羞", "脸红", "难为情", "社死", "丢脸", "出丑",
            "embarrassed", "awkward", "shy", "blush"
        ],
        "surprised": [
            "惊讶", "吃惊", "天啊", "哇塞", "哇", "居然", "竟然", "没想到", "出乎意料",
            "surprise", "wow", "omg", "oh my god", "amazing", "unbelievable"
        ],
        "shocked": [
            "震惊", "吓到", "惊呆了", "不敢相信", "震撼", "吓死", "恐怖", "害怕", "吓人",
            "shocked", "shocking", "scared", "frightened", "terrified", "horror"
        ],
        "thinking": [
            "思考", "考虑", "想一下", "琢磨", "沉思", "冥想", "想", "思考中", "在想",
            "think", "thinking", "consider", "ponder", "meditate"
        ],
        "winking": [
            "调皮", "眨眼", "你懂的", "坏笑", "邪恶", "奸笑", "使眼色",
            "wink", "teasing", "naughty", "mischievous"
        ],
        "cool": [
            "酷", "帅", "厉害", "棒极了", "真棒", "牛逼", "强", "优秀", "杰出", "出色",
            "cool", "awesome", "amazing", "great", "impressive", "perfect"
        ],
        "relaxed": [
            "放松", "舒服", "惬意", "悠闲", "轻松", "舒适", "安逸", "自在",
            "relax", "relaxed", "comfortable", "cozy", "chill", "peaceful"
        ],
        "delicious": [
            "好吃", "美味", "香", "馋", "可口", "香甜", "大餐", "大快朵颐", "流口水", "垂涎",
            "delicious", "yummy", "tasty", "yum", "appetizing", "mouthwatering"
        ],
        "kissy": [
            "亲亲", "么么", "吻", "mua", "muah", "亲一下", "飞吻",
            "kiss", "xoxo", "hug", "smooch"
        ],
        "confident": [
            "自信", "肯定", "确定", "毫无疑问", "当然", "必须的", "确信", "坚信",
            "confident", "sure", "certain", "definitely", "positive"
        ],
        "sleepy": [
            "困", "睡觉", "晚安", "想睡", "好累", "疲惫", "疲倦", "困了", "想休息", "睡意",
            "sleep", "sleepy", "tired", "exhausted", "bedtime", "good night"
        ],
        "silly": [
            "傻", "笨", "呆", "憨", "蠢", "二", "憨憨", "傻乎乎", "呆萌",
            "silly", "stupid", "dumb", "foolish", "goofy", "ridiculous"
        ],
        "confused": [
            "疑惑", "不明白", "不懂", "困惑", "疑问", "为什么", "怎么回事", "啥意思", "不清楚",
            "confused", "puzzled", "doubt", "question", "what", "why", "how"
        ]
    },
    "special_phrases": [
        {
            "phrases": [
                "你真", "你好", "您真", "你真棒", "你好厉害", "你太强了", "你真好", "你真聪明",
                "you are", "you're", "you look", "you seem", "so smart", "so kind"
            ],
            "emotion": "loving"
        },
        {
            "phrases": [
                "我真", "我最", "我太棒了", "我厉害", "我聪明", "我优秀",
                "i am", "i'm", "i feel", "so good", "so happy"
            ],
            "emotion": "cool"
        },
        {
            "phrases": [
                "睡觉", "晚安", "睡了", "好梦", "休息了", "去睡了",
                "sleep", "good night", "bedtime", "go to bed"
            ],
            "emotion": "sleepy"
        }
    ],
    "punctuation_rules": {
        "question_only": "thinking",
        "ellipsis": "thinking",
        "exclamation_positive": {
            "match_emotions": ["happy", "laughing", "cool"],
            "result": "laughing"
        },
        "exclamation_negative": {
            "match_emotions": ["angry", "sad", "crying"],
            "result": "angry"
        },
        "exclamation_default": "surprised"
    },
    "priority_order": [
        "laughing", "crying", "angry", "surprised", "shocked",
        "loving", "happy", "funny", "cool",
        "sad", "embarrassed", "confused",
        "thinking", "winking", "relaxed",
        "delicious", "kissy", "confident", "sleepy", "silly"
    ],
    "default_emotion": "happy",
    "long_text_threshold": 20,
    "long_text_repeat_weight": 0.5
}

JFB_emoji_mode

{
    "emotion_keywords": {
        "happy": [
            "开心", "高兴", "快乐", "愉快", "幸福", "满意", "棒", "好", "不错",
            "完美", "棒极了", "太好了", "好呀", "好的",
            "happy", "joy", "great", "good", "nice", "awesome", "fantastic", "wonderful"
        ],
        "laughing": [
            "哈哈", "哈哈哈", "呵呵", "嘿嘿", "嘻嘻", "笑死", "太好笑了", "笑死我了",
            "lol", "lmao", "haha", "hahaha", "hehe", "rofl", "funny", "laugh"
        ],
        "funny": [
            "搞笑", "滑稽", "逗", "幽默", "笑点", "段子", "笑话", "太逗了",
            "hilarious", "joke", "comedy"
        ],
        "sad": [
            "伤心", "难过", "悲哀", "悲伤", "忧郁", "郁闷", "沮丧", "失望",
            "想哭", "难受", "不开心", "唉", "呜呜",
            "sad", "upset", "unhappy", "depressed", "sorrow", "gloomy"
        ],
        "angry": [
            "生气", "愤怒", "气死", "讨厌", "烦人", "可恶", "烦死了", "恼火",
            "暴躁", "火大", "气炸了",
            "angry", "mad", "annoyed", "furious", "pissed", "hate"
        ],
        "crying": [
            "哭泣", "泪流", "大哭", "伤心欲绝", "泪目", "流泪", "哭死", "哭晕",
            "想哭", "泪崩",
            "cry", "crying", "tears", "sob", "weep"
        ],
        "loving": [
            "爱你", "喜欢", "爱", "亲爱的", "宝贝", "么么哒", "抱抱", "想你",
            "思念", "最爱", "亲亲", "喜欢你",
            "love", "like", "adore", "darling", "sweetie", "honey", "miss you", "heart"
        ],
        "embarrassed": [
            "尴尬", "不好意思", "害羞", "脸红", "难为情", "社死", "丢脸", "出丑",
            "embarrassed", "awkward", "shy", "blush"
        ],
        "surprised": [
            "惊讶", "吃惊", "天啊", "哇塞", "哇", "居然", "竟然", "没想到", "出乎意料",
            "surprise", "wow", "omg", "oh my god", "amazing", "unbelievable"
        ],
        "shocked": [
            "震惊", "吓到", "惊呆了", "不敢相信", "震撼", "吓死", "恐怖", "害怕", "吓人",
            "shocked", "shocking", "scared", "frightened", "terrified", "horror"
        ],
        "thinking": [
            "思考", "考虑", "想一下", "琢磨", "沉思", "冥想", "想", "思考中", "在想",
            "think", "thinking", "consider", "ponder", "meditate"
        ],
        "winking": [
            "调皮", "眨眼", "你懂的", "坏笑", "邪恶", "奸笑", "使眼色",
            "wink", "teasing", "naughty", "mischievous"
        ],
        "cool": [
            "酷", "帅", "厉害", "棒极了", "真棒", "牛逼", "强", "优秀", "杰出", "出色",
            "cool", "awesome", "amazing", "great", "impressive", "perfect"
        ],
        "relaxed": [
            "放松", "舒服", "惬意", "悠闲", "轻松", "舒适", "安逸", "自在",
            "relax", "relaxed", "comfortable", "cozy", "chill", "peaceful"
        ],
        "delicious": [
            "好吃", "美味", "香", "馋", "可口", "香甜", "大餐", "大快朵颐", "流口水", "垂涎",
            "delicious", "yummy", "tasty", "yum", "appetizing", "mouthwatering"
        ],
        "kissy": [
            "亲亲", "么么", "吻", "mua", "muah", "亲一下", "飞吻",
            "kiss", "xoxo", "hug", "smooch"
        ],
        "confident": [
            "自信", "肯定", "确定", "毫无疑问", "当然", "必须的", "确信", "坚信",
            "confident", "sure", "certain", "definitely", "positive"
        ],
        "sleepy": [
            "困", "睡觉", "晚安", "想睡", "好累", "疲惫", "疲倦", "困了", "想休息", "睡意",
            "sleep", "sleepy", "tired", "exhausted", "bedtime", "good night"
        ],
        "silly": [
            "傻", "笨", "呆", "憨", "蠢", "二", "憨憨", "傻乎乎", "呆萌",
            "silly", "stupid", "dumb", "foolish", "goofy", "ridiculous"
        ],
        "confused": [
            "疑惑", "不明白", "不懂", "困惑", "疑问", "为什么", "怎么回事", "啥意思", "不清楚",
            "confused", "puzzled", "doubt", "question", "what", "why", "how"
        ]
    },
    "special_phrases": [
        {
            "phrases": [
                "你真", "你好", "您真", "你真棒", "你好厉害", "你太强了", "你真好", "你真聪明",
                "you are", "you're", "you look", "you seem", "so smart", "so kind"
            ],
            "emotion": "loving"
        },
        {
            "phrases": [
                "我真", "我最", "我太棒了", "我厉害", "我聪明", "我优秀",
                "i am", "i'm", "i feel", "so good", "so happy"
            ],
            "emotion": "cool"
        },
        {
            "phrases": [
                "睡觉", "晚安", "睡了", "好梦", "休息了", "去睡了",
                "sleep", "good night", "bedtime", "go to bed"
            ],
            "emotion": "sleepy"
        }
    ],
    "punctuation_rules": {
        "question_only": "thinking",
        "ellipsis": "thinking",
        "exclamation_positive": {
            "match_emotions": ["happy", "laughing", "cool"],
            "result": "laughing"
        },
        "exclamation_negative": {
            "match_emotions": ["angry", "sad", "crying"],
            "result": "angry"
        },
        "exclamation_default": "surprised"
    },
    "priority_order": [
        "laughing", "crying", "angry", "surprised", "shocked",
        "loving", "happy", "funny", "cool",
        "sad", "embarrassed", "confused",
        "thinking", "winking", "relaxed",
        "delicious", "kissy", "confident", "sleepy", "silly"
    ],
    "default_emotion": "happy",
    "long_text_threshold": 20,
    "long_text_repeat_weight": 0.5
}

---

Important Notes

1. Performance Considerations

• ✅ Emoji matching is completed on the server side; client only needs to receive
• ✅ Matching algorithm is optimized with minimal performance impact (<1ms)
• ⚠️ Dimi mode sends more emoji messages; pay attention to message handling

2. Message Order

• 📌 EMOJI messages are sent immediately after the corresponding TEXT message
• 📌 One TEXT message may correspond to 0 or 1 EMOJI message
• 📌 The task_id of EMOJI messages matches the task_id of the current conversation

3. Error Handling

try:
    emoji_msg = parse_emoji_message(frame)
    if emoji_msg:
        show_emoji(emoji_msg['emoji_key'])
except Exception as e:
    logging.warning(f"Failed to parse emoji message: {e}")
    # Ignore error and continue processing other messages

4. Backward Compatibility

• If the client does not support EMOJI messages, it can directly ignore messages with msg_type == 9
• Does not affect normal receiving of other message types

---

Testing Suggestions

1. Functional Testing

# Test cases
test_cases = [
    ("false", "Hello", 0),                           # Expect 0 emojis
    ("true", "HAHAHA, so funny!", 1),               # Expect 1 emoji
    ("dimi", "Today is 666", 2),                    # Expect 2 emojis (question+answer)
]

for emoji_mode, input_text, expected_emoji_count in test_cases:
    # Send AUTH (set emoji_mode)
    # Send input
    # Count received EMOJI messages
    # Verify if it matches expectation

2. Stress Testing

• Test emoji message receiving under high concurrency
• Verify messages are not lost or out of order

3. Boundary Testing

• Empty text
• Very long text
• Special characters
• Multi-language mixing

---

FAQ

Q1: Why am I not receiving EMOJI messages?

A: Check the following:

1. Confirm that emoji_mode parameter was passed during authentication and is not false
2. Confirm that message parsing logic correctly handles msg_type == 9
3. Confirm buffer handling is correct without truncation

Q2: What is the order of EMOJI and TEXT messages?

A:

• TEXT message is sent first
• EMOJI message follows immediately after
• Order guarantee: TEXT → EMOJI → TEXT → EMOJI → ... → END_FRAME

Q3: Which mode should I choose - True / Dimi / Dimi2 / JFB?

A:

• True Mode: Suitable for general scenarios, 20 emotion expressions are sufficient to express basic emotions (answer-only matching)
• Dimi Mode: Suitable for specific scenarios, requires richer expressions (emoji sent in question phase includes playSound:1)
• Dimi2 Mode: Same behavior as Dimi, but question phase does not include playSound
• JFB Mode: Suitable for lean scenarios that only need 5 emotion categories (angry/sad/hesitant/anxious/excited), matches both question and answer, no playSound

Q4: How to customize emoji mapping?

A: Client maintains an emoji_key → animation/image mapping table:

{
  "happy": "happy_animation.anim",
  "laughing": "laugh_animation.anim",
  "xia_yu": "rain_icon.png"
}

Q5: What characters can emoji_key contain?

A:

• Contains only: lowercase letters (a-z), numbers (0-9), underscores (_)
• Examples: happy, xia_yu, tai_hao_xiao
• Maximum length: usually ≤ 20 characters

---

Related Documents

• V2.md - TCP Protocol Complete Documentation
• VAD Mode Integration Guide.md - VAD Mode Description
• mcp_protocal.md - MCP Tool Call Protocol

---

Changelog

v1.0 (2024-10-26)

• ✨ Initial version
• Added three emoji modes (false/true/dimi)
• Added EMOJI message type
• Supports both sentiment analysis and keyword matching algorithms

---

Technical Support

For questions, please contact the technical support team or refer to the complete protocol documentation.