nonebot-plugin-parser 2.0.14__tar.gz → 2.1.0__tar.gz

{nonebot_plugin_parser-2.0.14 → nonebot_plugin_parser-2.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: nonebot-plugin-parser
-Version: 2.0.14
+Version: 2.1.0
 Summary: NoneBot2 链接分享解析 Alconna 版, 通用媒体卡片渲染(PIL 实现), 支持 B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun/nga
 Keywords: nonebot,nonebot2,video,bilibili,youtube,tiktok,twitter,kuaishou,acfun,weibo,xiaohongshu,nga,douyin
 Author: fllesser

{nonebot_plugin_parser-2.0.14 → nonebot_plugin_parser-2.1.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "nonebot-plugin-parser"
-version = "2.0.14"
+version = "2.1.0"
 description = "NoneBot2 链接分享解析 Alconna 版, 通用媒体卡片渲染(PIL 实现), 支持 B站/抖音/快手/微博/小红书/youtube/tiktok/twitter/acfun/nga"
 authors = [{ "name" = "fllesser", "email" = "fllessive@gmail.com" }]
 readme = "README.md"
@@ -195,7 +195,7 @@ build-backend = "uv_build"
 
 
 [tool.bumpversion]
-current_version = "2.0.14"
+current_version = "2.1.0"
 commit = true
 message = "🔖 release: bump vesion from {current_version} to {new_version}"
 tag = true

{nonebot_plugin_parser-2.0.14 → nonebot_plugin_parser-2.1.0}/src/nonebot_plugin_parser/matchers/preprocess.py

@@ -4,7 +4,8 @@ from typing import Any, Literal
 
 from nonebot import logger
 from nonebot.matcher import Matcher
-from nonebot.message import event_preprocessor
+
+# from nonebot.message import event_preprocessor
 from nonebot.params import Depends
 from nonebot.plugin.on import get_matcher_source
 from nonebot.rule import Rule
@@ -42,30 +43,22 @@ def _kwd_regex_matched(state: T_State) -> re.Match[str] | None:
     return state.get(PSR_KWD_MATCHED_KEY)
 
 
-URL_KEY_MAPPING = {
-    "detail_1": "qqdocurl",
-    "news": "jumpUrl",
-    "music": "jumpUrl",
-}
-
-CHAR_REPLACEMENTS = {",": ",", "\\": "", "&": "&"}
-
-
-def _clean_url(url: str) -> str:
-    """清理 URL 中的特殊字符
-
+def _escape_raw(raw: str) -> str:
+    """
+    转义原始字符串中的特殊字符
     Args:
-        url: 原始 URL
+        raw: 原始字符串
 
     Returns:
-        str: 清理后的 URL
+        str: 转义后的字符串
     """
-    for old, new in CHAR_REPLACEMENTS.items():
-        url = url.replace(old, new)
-    return url
+    replacements = [(",", ","), ("\\", ""), ("&", "&")]
+    for old, new in replacements:
+        raw = raw.replace(old, new)
+    return raw
 
 
-def _extract_json_url(hyper: Hyper) -> str | None:
+def _extract_url(hyper: Hyper) -> str | None:
     """处理 JSON 类型的消息段，提取 URL
 
     Args:
@@ -75,12 +68,12 @@ def _extract_json_url(hyper: Hyper) -> str | None:
         Optional[str]: 提取的 URL, 如果提取失败则返回 None
     """
     data = hyper.data
-    raw_str = data.get("raw")
+    raw_str: str | None = data.get("raw")
 
     if raw_str is None:
         return None
-    # 处理转义字符
-    raw_str = raw_str.replace(",", ",")
+
+    raw_str = _escape_raw(raw_str)
 
     try:
         raw: dict[str, Any] = json.loads(raw_str)
@@ -92,24 +85,23 @@ def _extract_json_url(hyper: Hyper) -> str | None:
     if not meta:
         return None
 
-    for key1, key2 in URL_KEY_MAPPING.items():
-        if item := meta.get(key1):
-            if url := item.get(key2):
-                return _clean_url(url)
+    for key1, key2 in [("detail_1", "qqdocurl"), ("news", "jumpUrl"), ("music", "jumpUrl")]:
+        if url := meta.get(key1, {}).get(key2):
+            logger.debug(f"extract url from raw:{key1}:{key2}: {url}")
+            return url
     return None
 
 
-@event_preprocessor
-def extract_msg_text(message: UniMsg, state: T_State) -> None:
-    text: str | None = None
-
-    if hyper := message.get(Hyper, 1):
-        state[PSR_EXTRACT_KEY] = _extract_json_url(hyper.pop())
-        return
+# 纪念我写了一个存在了一年没人发现的 bug ()
+# @event_preprocessor
+# async def extract_msg_text(message: UniMsg, state: T_State):
+#     if hyper := next(iter(message.get(Hyper, 1)), None):
+#         state[PSR_EXTRACT_KEY] = _extract_url(hyper)
+#         return
 
-    # 提取纯文本
-    if text := message.extract_plain_text().strip():
-        state[PSR_EXTRACT_KEY] = text
+#     # 提取纯文本
+#     if text := message.extract_plain_text().strip():
+#         state[PSR_EXTRACT_KEY] = text
 
 
 class KeyPatternList(list[tuple[str, re.Pattern[str]]]):
@@ -138,14 +130,23 @@ class KeywordRegexRule:
     def __hash__(self) -> int:
         return hash(frozenset(self.key_pattern_list))
 
-    async def __call__(self, state: T_State, text: str = ExtractText()) -> bool:
+    async def __call__(self, message: UniMsg, state: T_State) -> bool:
+        text: str | None = None
+        if hyper := next(iter(message.get(Hyper, 1)), None):
+            text = _extract_url(hyper)
+
+        elif plain_text := message.extract_plain_text().strip():
+            text = plain_text
+
         if not text:
             return False
+
         for keyword, pattern in self.key_pattern_list:
             if keyword not in text:
                 continue
             if matched := pattern.search(text):
                 state[PSR_KWD_KEY] = keyword
+                state[PSR_EXTRACT_KEY] = text
                 state[PSR_KWD_MATCHED_KEY] = matched
                 return True
             logger.debug(f"keyword '{keyword}' is in '{text}', but not matched")

{nonebot_plugin_parser-2.0.14 → nonebot_plugin_parser-2.1.0}/src/nonebot_plugin_parser/parsers/bilibili/__init__.py

@@ -165,7 +165,7 @@ class BilibiliParser(BaseParser):
         # 判断链接类型并解析
         logger.debug(f"解析其他类型链接: {url}")
         # 1. 动态
-        if "t.bilibili.com" in url:
+        if "t.bili" in url or "m.bili" in url:
             return await self.parse_dynamic(url)
 
         # 2.图文动态