From 76f2124765368faa3f082ac7799d73a6cd74274a Mon Sep 17 00:00:00 2001
From: liuwei <liuwei@wdtrgf.com.cn>
Date: Thu, 23 Apr 2026 14:12:30 +0800
Subject: [PATCH] =?UTF-8?q?=E4=BF=AE=E5=A4=8D=E5=93=8D=E5=BA=94=E6=8C=87?=
 =?UTF-8?q?=E4=BB=A4=E8=AF=AD=E9=9F=B3=E5=8F=91=E9=80=81=E6=A0=BC=E5=BC=8F?=
 =?UTF-8?q?=E8=AF=AF=E5=88=A4=E5=AF=BC=E8=87=B4ffmpeg=E8=A7=A3=E7=A0=81?=
 =?UTF-8?q?=E5=A4=B1=E8=B4=A5?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

1. 语音发送逻辑改为优先按文件后缀推断格式，参考message_push.py的稳定实现。

2. 新增语音发送兜底重试机制：首选格式失败后自动尝试mp3/wav/amr。

3. 增加详细日志，便于排查配置格式与文件真实格式不一致问题。
---
 plugins/fun_command_play/main.py | 75 ++++++++++++++++++++++++++++++--
 1 file changed, 71 insertions(+), 4 deletions(-)

diff --git a/plugins/fun_command_play/main.py b/plugins/fun_command_play/main.py
index 01ff4f7..4e3a104 100644
--- a/plugins/fun_command_play/main.py
+++ b/plugins/fun_command_play/main.py
@@ -208,6 +208,70 @@ class FunCommandPlayPlugin(MessagePluginInterface):
             self._put_media_bytes_to_cache(cache_key, payload)
         return payload
 
+    @staticmethod
+    def _infer_voice_format_by_path(voice_path: str, configured_format: str) -> str:
+        """根据语音文件路径推断发送格式。
+
+        设计说明：
+        1. 参考 message_push.py 的既有稳定逻辑：优先按文件后缀判断格式。
+        2. 若后缀可识别（wav/mp3/amr），直接使用后缀，避免“配置格式与真实文件不一致”导致解码失败。
+        3. 若后缀不可识别，再回退到配置值；配置也无效时最终默认 mp3。
+        """
+        suffix = Path(str(voice_path or "")).suffix.lower().strip()
+        if suffix == ".wav":
+            return "wav"
+        if suffix == ".amr":
+            return "amr"
+        if suffix == ".mp3":
+            return "mp3"
+
+        normalized_cfg = str(configured_format or "").strip().lower()
+        if normalized_cfg in {"wav", "mp3", "amr"}:
+            return normalized_cfg
+        return "mp3"
+
+    async def _send_voice_with_fallback(
+        self,
+        bot: WechatAPIClient,
+        target_id: str,
+        voice_bytes: bytes,
+        voice_path: str,
+        configured_format: str,
+    ) -> None:
+        """发送语音并进行格式兜底重试。
+
+        背景：
+        - 在线上规则维护中，常见问题是“文件实际是 wav，但配置里误写 mp3”，
+          这会触发 ffmpeg 解码失败（Header missing）。
+
+        策略：
+        1. 首次发送：使用“后缀优先”推断格式（与 message_push.py 保持一致）。
+        2. 失败后自动尝试其它格式（mp3/wav/amr），提高容错性，降低人工维护成本。
+        """
+        first_try = self._infer_voice_format_by_path(voice_path=voice_path, configured_format=configured_format)
+        candidates: List[str] = [first_try]
+        for fallback in ("mp3", "wav", "amr"):
+            if fallback not in candidates:
+                candidates.append(fallback)
+
+        last_error: Optional[Exception] = None
+        for fmt in candidates:
+            try:
+                await bot.send_voice_message(target_id, voice_bytes, fmt)
+                if fmt != first_try:
+                    self.LOG.warning(
+                        f"[{self.name}] 语音发送触发格式兜底重试成功: path={voice_path}, first={first_try}, final={fmt}"
+                    )
+                return
+            except Exception as e:
+                last_error = e
+                self.LOG.warning(
+                    f"[{self.name}] 语音发送失败，尝试下一个格式: path={voice_path}, format={fmt}, error={e}"
+                )
+
+        if last_error:
+            raise last_error
+
     def start(self) -> bool:
         self.status = PluginStatus.RUNNING
         self.LOG.info(f"[{self.name}] 插件已启动")
@@ -425,10 +489,13 @@ class FunCommandPlayPlugin(MessagePluginInterface):
             voice_format = str(action.get("format", "") or "").strip().lower()
             voice_bytes = self._load_media_bytes(media_kind="voice", media_path=voice_path)
             if voice_bytes:
-                if not voice_format:
-                    suffix = Path(voice_path).suffix.lower()
-                    voice_format = "wav" if suffix == ".wav" else "mp3"
-                await bot.send_voice_message(target_id, voice_bytes, voice_format)
+                await self._send_voice_with_fallback(
+                    bot=bot,
+                    target_id=target_id,
+                    voice_bytes=voice_bytes,
+                    voice_path=voice_path,
+                    configured_format=voice_format,
+                )
             return
 
         if action_type == "video":