From a8a960a4ffe8b16258de93535dc033a12e0956a5 Mon Sep 17 00:00:00 2001
From: LoCCai <105721394+LoCCai@users.noreply.github.com>
Date: Sat, 13 Dec 2025 16:43:12 +0800
Subject: [PATCH] Improve emoji mixing functionality and message handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Refactor emoji mixing logic to support multiple emojis and deduplication. Enhance message handling for group chats with OneBot V11.

同时优化逻辑，直接进行多表情遍历合并发送，简化触发，方便使用。后续可以通过配置变量或者指令来设置多样化的触发方式和发送方式
---
 nonebot_plugin_emojimix/__init__.py | 122 +++++++++++++++++++++++-----
 1 file changed, 100 insertions(+), 22 deletions(-)

diff --git a/nonebot_plugin_emojimix/__init__.py b/nonebot_plugin_emojimix/__init__.py
index 7669f5d..39d1b84 100644
--- a/nonebot_plugin_emojimix/__init__.py
+++ b/nonebot_plugin_emojimix/__init__.py
@@ -1,11 +1,15 @@
 import re
+import base64
+from itertools import combinations
+from collections import Counter
 
 import emoji
-from nonebot import on_message, require
+from nonebot import on_message, require, Bot
 from nonebot.matcher import Matcher
 from nonebot.params import EventPlainText
 from nonebot.plugin import PluginMetadata, inherit_supported_adapters
 from nonebot.typing import T_State
+from nonebot.adapters import Event
 
 require("nonebot_plugin_alconna")
 
@@ -17,31 +21,36 @@
 __plugin_meta__ = PluginMetadata(
     name="emoji合成",
     description="将两个emoji合成为一张图片",
-    usage="{emoji1}+{emoji2}，如：😎+😁",
+    usage="发送两个及以上emoji，自动去重组合",
     type="application",
     homepage="https://github.com/noneplugin/nonebot-plugin-emojimix",
     config=Config,
     supported_adapters=inherit_supported_adapters("nonebot_plugin_alconna"),
     extra={
-        "example": "😎+😁",
+        "example": "😎😁🥺",
     },
 )
 
-
+# 过滤并生成 emoji 正则
 emojis = filter(lambda e: len(e) == 1, emoji.EMOJI_DATA.keys())
-emoji_pattern = "(" + "|".join(re.escape(e) for e in emojis) + ")"
-pattern = re.compile(
-    rf"^\s*(?P<code1>{emoji_pattern})\s*\+\s*(?P<code2>{emoji_pattern})\s*$"
-)
+# 构建匹配 pattern
+emoji_pattern_str = "|".join(re.escape(e) for e in emojis)
+# 编译正则
+emoji_regex = re.compile(f"({emoji_pattern_str})")
 
 
 async def check_eomjis(state: T_State, text: str = EventPlainText()) -> bool:
     text = text.strip()
-    if not text or "+" not in text:
+    if not text:
         return False
-    if matched := re.match(pattern, text):
-        state["code1"] = matched.group("code1")
-        state["code2"] = matched.group("code2")
+    
+    # 提取所有 emoji
+    matched = emoji_regex.findall(text)
+    valid_emojis = [e for e in matched if e.strip()]
+    
+    # 只要存在 2 个及以上 emoji 就触发处理逻辑
+    if len(valid_emojis) >= 2:
+        state["emoji_list"] = valid_emojis
         return True
     return False
 
@@ -50,13 +59,82 @@ async def check_eomjis(state: T_State, text: str = EventPlainText()) -> bool:
 
 
 @emojimix.handle()
-async def _(state: T_State, matcher: Matcher):
-    emoji_code1 = state["code1"]
-    emoji_code2 = state["code2"]
-
-    result = await mix_emoji(emoji_code1, emoji_code2)
-
-    if isinstance(result, str):
-        await matcher.finish(result)
-
-    await UniMessage.image(raw=result).send()
+async def _(bot: Bot, event: Event, state: T_State, matcher: Matcher):
+    raw_emoji_list = state["emoji_list"]
+    valid_results = []
+    
+    # --- 1. 数据预处理：去重与分类 ---
+    # 统计词频
+    counts = Counter(raw_emoji_list)
+    
+    # 找出所有重复出现的表情 (例如：['😎', '😎'] -> duplicate=['😎'])
+    duplicate_emojis = [e for e, count in counts.items() if count > 1]
+    
+    # 生成去重列表，使用 dict.fromkeys 保留原始输入顺序
+    unique_emojis = list(dict.fromkeys(raw_emoji_list))
+
+    # --- 2. 核心逻辑：自我合并测试 ---
+    # 如果一个表情出现了多次，说明用户可能想看它的自我融合（如 😳+😳）
+    for e in duplicate_emojis:
+        res = await mix_emoji(e, e)
+        # 只要不是字符串(错误信息)，就加入结果
+        if not isinstance(res, str):
+            valid_results.append(res)
+
+    # --- 3. 核心逻辑：交叉合并测试 ---
+    # 对去重后的列表进行两两组合 (C_n^2)，确保组合唯一性
+    # combinations 会生成迭代器：('A', 'B'), ('A', 'C'), ('B', 'C')
+    for e1, e2 in combinations(unique_emojis, 2):
+        res = await mix_emoji(e1, e2)
+        if not isinstance(res, str):
+            valid_results.append(res)
+
+    # --- 4. 结果校验 ---
+    # 如果没有生成任何一张图片（说明全是无法支持的组合），直接结束，不打扰群聊
+    if not valid_results:
+        await matcher.finish()
+
+    # --- 5. 消息构建与发送 ---
+    
+    # 尝试使用 OneBot V11 的合并转发
+    sent_via_forward = False
+    try:
+        from nonebot.adapters.onebot.v11 import Bot as V11Bot
+        from nonebot.adapters.onebot.v11 import MessageSegment as V11Seg
+        from nonebot.adapters.onebot.v11 import Message as V11Msg
+        
+        if isinstance(bot, V11Bot):
+            target_id = getattr(event, "group_id", None)
+            # 仅限群聊使用合并转发
+            if target_id:
+                nodes = []
+                for img_bytes in valid_results:
+                    b64_img = base64.b64encode(img_bytes).decode()
+                    file_uri = f"base64://{b64_img}"
+                    
+                    nodes.append(
+                        V11Seg.node_custom(
+                            user_id=int(bot.self_id),
+                            nickname="EmojiMix",
+                            content=V11Msg(V11Seg.image(file=file_uri))
+                        )
+                    )
+                
+                await bot.call_api("send_group_forward_msg", group_id=target_id, messages=nodes)
+                sent_via_forward = True
+
+    except Exception:
+        # 忽略所有 V11 相关错误，准备降级
+        pass
+
+    # 如果没有通过合并转发发送（非群聊、非V11、或报错），降级为普通消息发送
+    if not sent_via_forward:
+        msg = UniMessage()
+        for img_bytes in valid_results:
+            msg += UniMessage.image(raw=img_bytes)
+        # 直接发送
+        await msg.send()
+
+    # --- 6. 结束控制 ---
+    # 按照要求，最后使用 finish 空消息结束控制
+    await matcher.finish()