feat: 新增语音文件处理功能

app/asr/asr_voice_processor.py

@@ -117,10 +117,18 @@ class ASRVoiceProcessor:
         await redis_client_manager.close()
 
 
+import argparse
+
+
 async def main():
+    # parser = argparse.ArgumentParser(description='ASR语音文件处理')
+    # parser.add_argument('--batch-size', type=int, default=10, help='每批处理数量')
+    # parser.add_argument('--max-records', type=int, default=None, help='最大处理记录数，默认处理所有')
+    # args = parser.parse_args()
+
     processor = ASRVoiceProcessor()
     try:
-        await processor.process_voice_files(batch_size=10, max_records=10000)
+        await processor.process_voice_files(batch_size=10, max_records=0)
     finally:
         await processor.close()
 

app/models/mysql/archive_media_files.py

@@ -0,0 +1,50 @@
+import json
+from datetime import datetime
+
+from sqlalchemy import BigInteger, String, Text, Integer, TIMESTAMP, UniqueConstraint, Index
+from sqlalchemy.orm import mapped_column
+
+from app.models.mysql.archive_messages import Base
+
+
+class ArchiveMediaFiles(Base):
+    __tablename__ = 'archive_media_files'
+    __table_args__ = {
+        'comment': '企微会话存档媒体文件表',
+        'mysql_charset': 'utf8mb4',
+        'mysql_collate': 'utf8mb4_unicode_ci'
+    }
+
+    id = mapped_column(BigInteger, primary_key=True, autoincrement=True, comment='自增主键')
+    msg_id = mapped_column(String(255), nullable=False, comment='企微消息 msgid')
+    archive_message_id = mapped_column(BigInteger, nullable=False, comment='archive_messages.id')
+    sdkfileid = mapped_column(Text, nullable=True, comment='企微媒体文件 sdkfileid')
+    cos_url = mapped_column(String(1024), nullable=False, comment='COS 文件访问 URL')
+    file_size = mapped_column(BigInteger, default=0, comment='文件大小（字节）')
+    file_type = mapped_column(String(32), nullable=True, comment='文件类型：image/voice/video/file')
+    status = mapped_column(Integer, default=1, comment='状态：1-有效 0-无效')
+    created_at = mapped_column(TIMESTAMP, nullable=False, default=datetime.now, comment='创建时间')
+    updated_at = mapped_column(TIMESTAMP, nullable=False, default=datetime.now, onupdate=datetime.now, comment='更新时间')
+
+    __table_args__ = (
+        UniqueConstraint('archive_message_id', name='uk_archive_msg_id'),
+        UniqueConstraint('msg_id', name='uk_msg_id'),
+        Index('idx_status', 'status'),
+    )
+
+    def to_dict(self):
+        return {
+            'id': self.id,
+            'msg_id': self.msg_id,
+            'archive_message_id': self.archive_message_id,
+            'sdkfileid': self.sdkfileid,
+            'cos_url': self.cos_url,
+            'file_size': self.file_size,
+            'file_type': self.file_type,
+            'status': self.status,
+            'created_at': self.created_at.isoformat() if self.created_at else None,
+            'updated_at': self.updated_at.isoformat() if self.updated_at else None
+        }
+
+    def __repr__(self):
+        return json.dumps(self.to_dict())

app/repository/archive_media_files_repository.py

@@ -42,12 +42,13 @@ class ArchiveMediaFilesRepository:
         result = await self.session.execute(query)
         return result.scalar() or 0
 
-    async def get_by_archive_message_id(self, archive_message_id: int) -> Optional[ArchiveMediaFiles]:
+    async def get_by_archive_message_id(self, archive_message_id: int, file_type: str = "voice") -> Optional[ArchiveMediaFiles]:
         """
         根据 archive_message_id 查询媒体文件
 
         Args:
             archive_message_id: 归档消息ID
+            file_type: 文件类型过滤，默认 voice
 
         Returns:
             ArchiveMediaFiles 对象或 None
@@ -56,5 +57,8 @@ class ArchiveMediaFilesRepository:
             ArchiveMediaFiles.archive_message_id == archive_message_id
         )
 
+        if file_type:
+            query = query.where(ArchiveMediaFiles.file_type == file_type)
+
         result = await self.session.execute(query)
         return result.scalar_one_or_none()