tingwu

backend/ai_services/views.py

@@ -0,0 +1,140 @@
+import logging
+import uuid
+from rest_framework import viewsets, status
+from rest_framework.decorators import action
+from rest_framework.response import Response
+from rest_framework.parsers import MultiPartParser, FormParser
+from django.conf import settings
+from .models import TranscriptionTask
+from .serializers import TranscriptionTaskSerializer
+from .services import AliyunTingwuService
+
+logger = logging.getLogger(__name__)
+
+class TranscriptionTaskViewSet(viewsets.ModelViewSet):
+    queryset = TranscriptionTask.objects.all()
+    serializer_class = TranscriptionTaskSerializer
+    parser_classes = (MultiPartParser, FormParser)
+
+    def create(self, request, *args, **kwargs):
+        """
+        上传音频文件并创建听悟转写任务
+        """
+        file_obj = request.FILES.get('file')
+        if not file_obj:
+            return Response({'error': '未提供文件'}, status=status.HTTP_400_BAD_REQUEST)
+
+        service = AliyunTingwuService()
+        if not service.bucket or not service.client:
+            return Response({'error': '阿里云服务未配置'}, status=status.HTTP_503_SERVICE_UNAVAILABLE)
+
+        try:
+            # 1. 上传文件到 OSS
+            file_extension = file_obj.name.split('.')[-1]
+            file_name = f"transcription/{uuid.uuid4()}.{file_extension}"
+            
+            # 使用服务上传
+            oss_url = service.upload_to_oss(file_obj, file_name)
+            
+            # 2. 创建数据库记录
+            task_record = TranscriptionTask.objects.create(
+                file_url=oss_url,
+                status=TranscriptionTask.Status.PENDING
+            )
+
+            # 3. 调用听悟接口创建任务
+            try:
+                tingwu_response = service.create_transcription_task(oss_url)
+                task_id = tingwu_response.get('TaskId')
+                
+                if task_id:
+                    task_record.task_id = task_id
+                    task_record.status = TranscriptionTask.Status.PROCESSING
+                    task_record.save()
+                else:
+                    task_record.status = TranscriptionTask.Status.FAILED
+                    task_record.error_message = "未能获取 TaskId"
+                    task_record.save()
+                    return Response({'error': '未能获取 TaskId'}, status=status.HTTP_500_INTERNAL_SERVER_ERROR)
+
+            except Exception as e:
+                task_record.status = TranscriptionTask.Status.FAILED
+                task_record.error_message = str(e)
+                task_record.save()
+                logger.error(f"创建听悟任务失败: {e}")
+                return Response({'error': f"创建听悟任务失败: {str(e)}"}, status=status.HTTP_500_INTERNAL_SERVER_ERROR)
+
+            serializer = self.get_serializer(task_record)
+            return Response(serializer.data, status=status.HTTP_201_CREATED)
+
+        except Exception as e:
+            logger.error(f"处理上传请求失败: {e}")
+            return Response({'error': str(e)}, status=status.HTTP_500_INTERNAL_SERVER_ERROR)
+
+    @action(detail=True, methods=['get'])
+    def refresh_status(self, request, pk=None):
+        """
+        刷新任务状态并获取结果
+        """
+        task = self.get_object()
+        
+        # 如果任务已经完成或失败，直接返回当前状态
+        if task.status in [TranscriptionTask.Status.SUCCEEDED, TranscriptionTask.Status.FAILED]:
+            serializer = self.get_serializer(task)
+            return Response(serializer.data)
+
+        if not task.task_id:
+            return Response({'error': '任务ID不存在'}, status=status.HTTP_400_BAD_REQUEST)
+
+        service = AliyunTingwuService()
+        try:
+            result = service.get_task_info(task.task_id)
+            task_status = result.get('TaskStatus')
+            
+            if task_status == 'COMPLETE':
+                task.status = TranscriptionTask.Status.SUCCEEDED
+                
+                # 解析结果
+                task_result = result.get('Result', {})
+                
+                # 提取逐字稿
+                sentences = task_result.get('Transcription', {}).get('Sentences', [])
+                full_text = " ".join([s.get('Text', '') for s in sentences])
+                task.transcription = full_text
+                
+                # 提取总结
+                # 总结结果结构可能因配置不同而异，这里尝试获取摘要
+                summarization = task_result.get('Summarization', {})
+                # 听悟的总结通常在 Summarization.Text 或类似字段
+                # 如果是章节摘要，可能在 Chapters 中
+                # 假设是全文摘要
+                if 'Text' in summarization:
+                    task.summary = summarization['Text']
+                elif 'Headline' in summarization:
+                    task.summary = summarization['Headline']
+                else:
+                    # 尝试从章节摘要中提取
+                    chapters = task_result.get('Chapters', [])
+                    summary_parts = []
+                    for chapter in chapters:
+                        if 'Headline' in chapter:
+                            summary_parts.append(chapter['Headline'])
+                        if 'Summary' in chapter:
+                            summary_parts.append(chapter['Summary'])
+                    task.summary = "\n".join(summary_parts)
+
+                task.save()
+
+            elif task_status == 'FAILED':
+                task.status = TranscriptionTask.Status.FAILED
+                task.error_message = result.get('TaskStatusText', 'Unknown error')
+                task.save()
+            
+            # 其他状态 (PENDING, RUNNING) 不做更改
+            
+            serializer = self.get_serializer(task)
+            return Response(serializer.data)
+            
+        except Exception as e:
+            logger.error(f"刷新任务状态失败: {e}")
+            return Response({'error': str(e)}, status=status.HTTP_500_INTERNAL_SERVER_ERROR)