重构匹配逻辑 (#22)

1. 添加事件预处理，用于提取小程序链接
2. 优化 链接/资源ID 提取逻辑

nonebot_plugin_resolver2/__init__.py

@@ -1,7 +1,7 @@
 from nonebot import get_driver, logger
 
 from nonebot.plugin import PluginMetadata
-from .matchers import resolvers, commands
+from .matchers import resolvers
 from .config import *
 from .cookie import *
 

nonebot_plugin_resolver2/config.py

@@ -1,4 +1,8 @@
-from nonebot import get_driver, require, get_plugin_config
+from nonebot import (
+    get_driver, 
+    require, 
+    get_plugin_config
+)
 require("nonebot_plugin_localstore")
 require("nonebot_plugin_apscheduler")
 from nonebot_plugin_apscheduler import scheduler
@@ -30,6 +34,6 @@ class Config(BaseModel):
 # 全局名称
 NICKNAME: str = next(iter(get_driver().config.nickname), "")
 # 根据是否为国外机器声明代理
-PROXY: str = None if rconfig.r_is_oversea else rconfig.r_proxy
+PROXY: str = "" if rconfig.r_is_oversea else rconfig.r_proxy
 # 哔哩哔哩限制的最大视频时长（默认8分钟）单位：秒
 DURATION_MAXIMUM: int = rconfig.r_video_duration_maximum

nonebot_plugin_resolver2/matchers/__init__.py

@@ -5,5 +5,5 @@
 
 resolvers = {module: eval(module) for module in modules}
 
-from .filter import *
-commands = [enable_resolve, disable_resolve, check_resolve, enable_all_resolve, disable_all_resolve]
+# from .filter import *
+# commands = [enable_resolve, disable_resolve, check_resolve, enable_all_resolve, disable_all_resolve]

nonebot_plugin_resolver2/matchers/bilibili.py

@@ -4,7 +4,7 @@
 
 from tqdm.asyncio import tqdm
 from nonebot.log import logger
-from nonebot.rule import Rule
+from nonebot.typing import T_State
 from nonebot.params import CommandArg
 from nonebot.exception import ActionFailed
 from nonebot.plugin.on import on_message, on_command
@@ -31,7 +31,12 @@
     get_file_seg
 )
 from .filter import is_not_in_disable_group
-from ..data_source.common import (
+from .preprocess import (
+    r_keywords,
+    R_KEYWORD_KEY,
+    R_EXTRACT_KEY
+)
+from ..download.common import (
     delete_boring_characters,
     download_file_by_stream,
     merge_av
@@ -54,52 +59,43 @@
     'referer': 'https://www.bilibili.com'
 }
 
-def is_bilibili(event: MessageEvent) -> bool:
-    message = str(event.message).strip()
-    return any(key in message for key in {"bilibili.com", "b23.tv", "bili2233.cn", "BV"})
-
 bilibili = on_message(
-    rule = Rule(
-        is_not_in_disable_group,
-        is_bilibili
-    )
+    rule = is_not_in_disable_group & r_keywords("bilibili", "b23", "bili2233", "BV")
 )
+
 bili_music = on_command(
     cmd="bm",
     block = True
 )
 
 @bilibili.handle()
-async def _(bot: Bot, event: MessageEvent):
+async def _(bot: Bot, state: T_State):
     # 消息
-    message: str = str(event.message).strip()
+    text, keyword = state.get(R_EXTRACT_KEY), state.get(R_KEYWORD_KEY)
+    url, video_id = '', ''
 
-    url: str = ""
-    video_id: str = ""
-    # BV处理
-    if re.match(r'^BV[1-9a-zA-Z]{10}$', message):
-        # url = 'https://www.bilibili.com/video/' + message
-        video_id = message
-    # 处理短号、小程序问题
-    elif 'b23.tv' in message:
+    if keyword == 'BV':
+        if re.match(r'^BV[1-9a-zA-Z]{10}$', text):
+            video_id = text
+    elif keyword == 'b23':
+        # 处理短号、小程序
         b_short_reg = r"(http:|https:)\/\/b23.tv\/[A-Za-z\d._?%&+\-=\/#]*"
-        if match := re.search(b_short_reg, message.replace("\\", "")):
+        if match := re.search(b_short_reg, text):
             b_short_url = match.group(0)
             async with httpx.AsyncClient() as client:
                 resp = await client.get(b_short_url, headers=BILIBILI_HEADERS, follow_redirects=True)
             url = str(resp.url)
-    elif 'bili2233' in message:
+    elif keyword == 'bili2233':
+        # 处理新域名、小程序
         b_new_reg = r"(http:|https:)\/\/bili2233.cn\/[A-Za-z\d._?%&+\-=\/#]*"
-        #await bilibili.send(message.replace("\\", ""))
-        if match := re.search(b_new_reg, message.replace("\\", "")):
+        if match := re.search(b_new_reg, text):
             b_new_url = match.group(0)
             async with httpx.AsyncClient() as client:
                 resp = await client.get(b_new_url, headers=BILIBILI_HEADERS, follow_redirects=True)
             url = str(resp.url)
-
     else:
-        url_reg = r"(http:|https:)\/\/(space|www|live).bilibili.com\/[A-Za-z\d._?%&+\-=\/#]*"
-        if match := re.search(url_reg, message):
+        url_reg = r"(http:|https:)\/\/(space|www|live|m)?.?bilibili.com\/[A-Za-z\d._?%&+\-=\/#]*"
+        if match := re.search(url_reg, text):
             url = match.group(0)
     if url:
         # ===============发现解析的是动态，转移一下===============
@@ -174,8 +170,8 @@ async def _(bot: Bot, event: MessageEvent):
 
     if video_id:
         v = video.Video(bvid = video_id, credential=credential)
-    elif match := re.search(r"video\/[^\?\/ ]+", url):
-        video_id = match.group(0).split('/')[1]
+    elif match := re.search(r"(av\d+|BV[A-Za-z0-9]{10})", url):
+        video_id = match.group(1)
         if "av" in video_id:
             v = video.Video(aid=int(video_id.split("av")[1]), credential=credential)
         else:
@@ -186,11 +182,11 @@ async def _(bot: Bot, event: MessageEvent):
     segs: list[MessageSegment | str] = []
     try:
         video_info = await v.get_info()
-        if video_info is None:
-            await bilibili.finish(Message(f"{NICKNAME}解析 | 哔哩哔哩 - 出错，无法获取数据！"))
-        await bilibili.send(f'{NICKNAME}解析 | 哔哩哔哩 - 视频')
+        if not video_info:
+            raise Exception("video_info is None")
     except Exception as e:
-        await bilibili.finish(Message(f"{NICKNAME}解析 | 哔哩哔哩 - 出错\n{e}"))
+        await bilibili.finish(f"{NICKNAME}解析 | 哔哩哔哩 - 出错 {e}")
+    await bilibili.send(f'{NICKNAME}解析 | 哔哩哔哩 - 视频')
     video_title, video_cover, video_desc, video_duration = video_info['title'], video_info['pic'], video_info['desc'], video_info['duration']
     # 校准 分 p 的情况
     page_num = 0

nonebot_plugin_resolver2/matchers/douyin.py

@@ -14,7 +14,7 @@
 from .utils import get_video_seg, construct_nodes
 from .filter import is_not_in_disable_group
 
-from ..data_source.common import download_img
+from ..download.common import download_img
 from ..parsers.base import VideoInfo
 from ..parsers.douyin import DouYin
 from ..config import NICKNAME

nonebot_plugin_resolver2/matchers/kugou.py

@@ -4,8 +4,9 @@
 import httpx
 import asyncio
 
-from nonebot import on_keyword
-from nonebot.rule import Rule
+from nonebot.log import logger
+from nonebot.typing import T_State
+from nonebot.plugin import on_message
 from nonebot.adapters.onebot.v11 import (
     Message,
     MessageEvent,
@@ -14,52 +15,41 @@
 )
 from .utils import get_file_seg
 from .filter import is_not_in_disable_group
-
-from ..data_source.common import download_audio
+from .preprocess import (
+    r_keywords,
+    R_KEYWORD_KEY,
+    R_EXTRACT_KEY
+)
+from ..download.common import download_audio
 from ..constant import COMMON_HEADER
 from ..config import NICKNAME
 
 # KG临时接口
 KUGOU_TEMP_API = "https://www.hhlqilongzhu.cn/api/dg_kugouSQ.php?msg={}&n=1&type=json"
 
-kugou = on_keyword(
-    keywords = {"kugou.com"},
-    rule = Rule(is_not_in_disable_group)
+kugou = on_message(
+    rule = is_not_in_disable_group & r_keywords("kugou.com")
 )
 
 @kugou.handle()
-async def _(bot: Bot, event: MessageEvent):
-    message = event.message.extract_plain_text().strip()
-    # logger.info(message)
-    reg1 = r"https?://.*?kugou\.com.*?(?=\s|$|\n)"
-    reg2 = r'jumpUrl":\s*"(https?:\\/\\/[^"]+)"'
-    reg3 = r'jumpUrl":\s*"(https?://[^"]+)"'
+async def _(bot: Bot, state: T_State):
+    text = state.get(R_EXTRACT_KEY)
+    pattern = r"https?://.*?kugou\.com.*?(?=\s|$|\n)"
     # 处理卡片问题
-    if 'com.tencent.structmsg' in message:
-        if match := re.search(reg2, message):
-            get_url = match.group(1)
-        else:
-            if match := re.search(reg3, message):
-                get_url = match.group(1)
-            else:
-                await kugou.send(Message(f"{NICKNAME}解析 | 酷狗音乐 - 获取链接失败"))
-                get_url = None
-                return
-        if get_url:
-            url = json.loads('"' + get_url + '"')
+    if match := re.search(pattern, text):
+        url = match.group(0)
     else:
-        match = re.search(reg1, message)
-        url = match.group()
-
-        # 使用 httpx 获取 URL 的标题
+        logger.info(f'无效链接，忽略 - {text}')
+        return
+    # 使用 httpx 获取 URL 的标题
     async with httpx.AsyncClient() as client:
         response = await client.get(url, follow_redirects=True)
     if response.status_code != 200:
         await kugou.finish(f"{NICKNAME}解析 | 酷狗音乐 - 获取链接失败")
     title = response.text
-    get_name = r"<title>(.*?)_高音质在线试听"
-    if name := re.search(get_name, title):
-        kugou_title = name.group(1)  # 只输出歌曲名和歌手名的部分
+    pattern = r"<title>(.*?)_高音质在线试听"
+    if match := re.search(pattern, title):
+        kugou_title = match.group(1)  # 只输出歌曲名和歌手名的部分
         async with httpx.AsyncClient() as client:
             resp = await client.get(f"{KUGOU_TEMP_API.replace('{}', kugou_title)}", headers=COMMON_HEADER)
             kugou_vip_data = resp.json()
@@ -80,5 +70,3 @@ async def _(bot: Bot, event: MessageEvent):
         await kugou.finish(get_file_seg(audio_path, f'{kugou_name}-{kugou_singer}.{audio_path.name.split(".")[-1]}'))
     else:
         await kugou.send(f"{NICKNAME}解析 | 酷狗音乐 - 不支持当前外链，请重新分享再试")
-
-

nonebot_plugin_resolver2/matchers/ncm.py

@@ -2,19 +2,23 @@
 import httpx
 import asyncio
 
-from nonebot import on_message
-from nonebot.rule import Rule
+from nonebot.plugin import on_message
+from nonebot.typing import T_State
 from nonebot.adapters.onebot.v11 import (
     Message,
-    MessageEvent,
     Bot,
     MessageSegment
 )
 
 from .filter import is_not_in_disable_group
 from .utils import get_file_seg
+from .preprocess import (
+    r_keywords,
+    R_KEYWORD_KEY,
+    R_EXTRACT_KEY
+)
 from ..constant import COMMON_HEADER
-from ..data_source.common import download_audio
+from ..download.common import download_audio
 from ..config import *
 
 # NCM获取歌曲信息链接
@@ -23,27 +27,23 @@
 # NCM临时接口
 NETEASE_TEMP_API = "https://www.hhlqilongzhu.cn/api/dg_wyymusic.php?id={}&br=7&type=json"
 
-def is_ncm(event: MessageEvent) -> bool:
-    message = str(event.message).strip()
-    return any(key in message for key in {"music.163.com", "163cn.tv"})
-
 ncm = on_message(
-    rule = Rule(is_ncm, is_not_in_disable_group)
+    rule = is_not_in_disable_group & r_keywords("music.163.com", "163cn.tv")
 )
 
 @ncm.handle()
-async def ncm_handler(bot: Bot, event: MessageEvent):
-    message = str(event.message).strip()
+async def _(bot: Bot, state: T_State):
+    text, keyword = state.get(R_EXTRACT_KEY), state.get(R_KEYWORD_KEY)
     # 解析短链接
-    if "163cn.tv" in message:
-        if match := re.search(r"(http:|https:)\/\/163cn\.tv\/([a-zA-Z0-9]+)", message):
-            message = match.group(0)
-        # message = str(httpx.head(message, follow_redirects=True).url)
+    if keyword == "163cn.tv":
+        if match := re.search(r"(http:|https:)\/\/163cn\.tv\/([a-zA-Z0-9]+)", text):
+            url = match.group(0)
         async with httpx.AsyncClient() as client:
-            resp = await client.head(message, follow_redirects=True)
-            message = str(resp.url)
-
-    if match := re.search(r"id=(\d+)", message):
+            resp = await client.head(url, follow_redirects=True)
+        url = str(resp.url)
+    else:
+        url = text
+    if match := re.search(r"id=(\d+)", url):
         ncm_id = match.group(1)
     else:
         await ncm.finish(f"{NICKNAME}解析 | 网易云 - 获取链接失败")

nonebot_plugin_resolver2/matchers/preprocess.py

@@ -0,0 +1,63 @@
+import json
+
+from typing import Literal
+from nonebot.log import logger
+from nonebot.message import event_preprocessor
+from nonebot.typing import T_State
+from nonebot.rule import Rule
+from nonebot.adapters.onebot.v11 import MessageEvent
+
+
+R_KEYWORD_KEY: Literal["_r_keyword"] = "_r_keyword"
+R_EXTRACT_KEY: Literal["_r_extract"] = "_r_extract"
+
+@event_preprocessor
+def _(event: MessageEvent, state: T_State): 
+    message = event.get_message()
+    text = message.extract_plain_text().strip()
+    if json_seg := [seg for seg in message if seg.type == 'json']:
+        try:
+            data_str = json_seg[0].data.get('data')
+            data_str = data_str.replace(',', ',')
+            data = json.loads(data_str)
+            meta = data.get('meta')
+            if detail := meta.get('detail_1'):
+                text = detail.get('qqdocurl')
+            elif news := meta.get('news'):
+                text = news.get('jumpUrl')
+            text = text.replace('\\', '').replace("&", "&")
+        except Exception:
+            pass
+    state[R_EXTRACT_KEY] = text
+
+class RKeywordsRule:
+    """检查消息是否含有关键词 增强版"""
+
+    __slots__ = ("keywords",)
+
+    def __init__(self, *keywords: str):
+        self.keywords = keywords
+
+    def __repr__(self) -> str:
+        return f"RKeywords(keywords={self.keywords})"
+
+    def __eq__(self, other: object) -> bool:
+        return isinstance(other, RKeywordsRule) and frozenset(
+            self.keywords
+        ) == frozenset(other.keywords)
+
+    def __hash__(self) -> int:
+        return hash(frozenset(self.keywords))
+
+    async def __call__(self, state: T_State) -> bool:
+        text = state.get(R_EXTRACT_KEY)
+        if not text:
+            return False
+        if key := next((k for k in self.keywords if k in text), None):
+            state[R_KEYWORD_KEY] = key
+            return True
+        return False
+
+
+def r_keywords(*keywords: str) -> Rule:
+    return Rule(RKeywordsRule(*keywords))

nonebot_plugin_resolver2/matchers/tiktok.py

@@ -10,7 +10,7 @@
 
 from .filter import is_not_in_disable_group
 from .utils import get_video_seg
-from ..data_source.ytdlp import get_video_info, ytdlp_download_video
+from ..download.ytdlp import get_video_info, ytdlp_download_video
 from ..config import *