276 lines
8.7 KiB
Python
276 lines
8.7 KiB
Python
"""
|
|
切片Repository
|
|
提供切片相关的数据访问操作
|
|
"""
|
|
|
|
from typing import List, Optional, Dict, Any
|
|
from sqlalchemy.orm import Session
|
|
from sqlalchemy import desc, asc, func
|
|
from pathlib import Path
|
|
from .base import BaseRepository
|
|
from models.clip import Clip, ClipStatus
|
|
|
|
class ClipRepository(BaseRepository[Clip]):
|
|
"""切片Repository类"""
|
|
|
|
def __init__(self, db: Session):
|
|
super().__init__(Clip, db)
|
|
|
|
def get_by_project(self, project_id: str) -> List[Clip]:
|
|
"""
|
|
获取项目的所有切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.find_by(project_id=project_id)
|
|
|
|
def get_by_status(self, status: ClipStatus) -> List[Clip]:
|
|
"""
|
|
根据状态获取切片列表
|
|
|
|
Args:
|
|
status: 切片状态
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.find_by(status=status)
|
|
|
|
def get_by_project_and_status(self, project_id: str, status: ClipStatus) -> List[Clip]:
|
|
"""
|
|
根据项目和状态获取切片列表
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
status: 切片状态
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.find_by(project_id=project_id, status=status)
|
|
|
|
def get_high_score_clips(self, project_id: str, min_score: float = 0.7, limit: int = 10) -> List[Clip]:
|
|
"""
|
|
获取高分切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
min_score: 最低评分
|
|
limit: 返回数量限制
|
|
|
|
Returns:
|
|
高分切片列表
|
|
"""
|
|
return self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.score >= min_score
|
|
).order_by(desc(self.model.score)).limit(limit).all()
|
|
|
|
def get_clips_by_duration_range(self, project_id: str, min_duration: int, max_duration: int) -> List[Clip]:
|
|
"""
|
|
根据时长范围获取切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
min_duration: 最小时长（秒）
|
|
max_duration: 最大时长（秒）
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.duration >= min_duration,
|
|
self.model.duration <= max_duration
|
|
).order_by(asc(self.model.start_time)).all()
|
|
|
|
def get_clips_by_time_range(self, project_id: str, start_time: int, end_time: int) -> List[Clip]:
|
|
"""
|
|
根据时间范围获取切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
start_time: 开始时间（秒）
|
|
end_time: 结束时间（秒）
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.start_time >= start_time,
|
|
self.model.end_time <= end_time
|
|
).order_by(asc(self.model.start_time)).all()
|
|
|
|
def create_clip(self, clip_data: Dict[str, Any]) -> Clip:
|
|
"""创建切片记录（分离存储模式）"""
|
|
from services.storage_service import StorageService
|
|
import uuid
|
|
|
|
# 生成切片ID（如果没有提供）
|
|
if "id" not in clip_data:
|
|
clip_data["id"] = str(uuid.uuid4())
|
|
|
|
# 1. 保存切片文件到文件系统
|
|
storage_service = StorageService(clip_data["project_id"])
|
|
video_path = storage_service.save_clip_file(clip_data, clip_data["id"])
|
|
|
|
# 2. 保存完整数据到文件系统
|
|
metadata_path = storage_service.save_metadata(clip_data, f"clip_{clip_data['id']}")
|
|
|
|
# 3. 保存元数据到数据库（只存储路径引用）
|
|
clip = Clip(
|
|
id=clip_data["id"],
|
|
project_id=clip_data["project_id"],
|
|
title=clip_data["title"],
|
|
description=clip_data.get("description"),
|
|
start_time=clip_data["start_time"],
|
|
end_time=clip_data["end_time"],
|
|
duration=clip_data["duration"],
|
|
score=clip_data.get("score"),
|
|
video_path=video_path, # 只存储路径
|
|
clip_metadata={
|
|
'metadata_file': metadata_path, # 完整数据文件路径
|
|
'clip_id': clip_data["id"],
|
|
'created_at': clip_data.get("created_at")
|
|
}
|
|
)
|
|
|
|
self.db.add(clip)
|
|
self.db.commit()
|
|
return clip
|
|
|
|
def get_clip_file(self, clip_id: str) -> Optional[Path]:
|
|
"""获取切片文件路径"""
|
|
clip = self.get_by_id(clip_id)
|
|
if clip and clip.video_path:
|
|
return Path(clip.video_path)
|
|
return None
|
|
|
|
def get_clip_content(self, clip_id: str) -> Optional[Dict[str, Any]]:
|
|
"""获取切片完整内容"""
|
|
clip = self.get_by_id(clip_id)
|
|
if not clip:
|
|
return None
|
|
|
|
# 从文件系统获取完整数据
|
|
if clip.clip_metadata and 'metadata_file' in clip.clip_metadata:
|
|
from services.storage_service import StorageService
|
|
storage_service = StorageService(clip.project_id)
|
|
return storage_service.get_file_content(clip.clip_metadata['metadata_file'])
|
|
|
|
return None
|
|
|
|
def search_clips(self, project_id: str, keyword: str) -> List[Clip]:
|
|
"""
|
|
搜索切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
keyword: 搜索关键词
|
|
|
|
Returns:
|
|
匹配的切片列表
|
|
"""
|
|
return self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
(self.model.title.contains(keyword) |
|
|
self.model.description.contains(keyword) |
|
|
self.model.recommendation_reason.contains(keyword))
|
|
).all()
|
|
|
|
def get_clips_statistics(self, project_id: str) -> dict:
|
|
"""
|
|
获取切片统计信息
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
|
|
Returns:
|
|
统计信息字典
|
|
"""
|
|
total_clips = self.db.query(self.model).filter(
|
|
self.model.project_id == project_id
|
|
).count()
|
|
|
|
completed_clips = self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.status == ClipStatus.COMPLETED
|
|
).count()
|
|
|
|
avg_score = self.db.query(func.avg(self.model.score)).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.score.isnot(None)
|
|
).scalar()
|
|
|
|
total_duration = self.db.query(func.sum(self.model.duration)).filter(
|
|
self.model.project_id == project_id
|
|
).scalar()
|
|
|
|
return {
|
|
"total": total_clips,
|
|
"completed": completed_clips,
|
|
"avg_score": float(avg_score) if avg_score else 0.0,
|
|
"total_duration": int(total_duration) if total_duration else 0,
|
|
"completion_rate": (completed_clips / total_clips * 100) if total_clips > 0 else 0
|
|
}
|
|
|
|
def update_clip_status(self, clip_id: str, status: ClipStatus) -> Optional[Clip]:
|
|
"""
|
|
更新切片状态
|
|
|
|
Args:
|
|
clip_id: 切片ID
|
|
status: 新状态
|
|
|
|
Returns:
|
|
更新后的切片实例或None
|
|
"""
|
|
return self.update(clip_id, status=status)
|
|
|
|
def update_clip_score(self, clip_id: str, score: float) -> Optional[Clip]:
|
|
"""
|
|
更新切片评分
|
|
|
|
Args:
|
|
clip_id: 切片ID
|
|
score: 新评分
|
|
|
|
Returns:
|
|
更新后的切片实例或None
|
|
"""
|
|
return self.update(clip_id, score=score)
|
|
|
|
def get_clips_for_collection(self, project_id: str, collection_size: int = 5) -> List[Clip]:
|
|
"""
|
|
获取适合合集的切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
collection_size: 合集大小
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.db.query(self.model).filter(
|
|
self.model.project_id == project_id,
|
|
self.model.status == ClipStatus.COMPLETED,
|
|
self.model.score >= 0.7
|
|
).order_by(desc(self.model.score)).limit(collection_size).all()
|
|
|
|
def get_clips_by_processing_step(self, project_id: str, step: int) -> List[Clip]:
|
|
"""
|
|
根据处理步骤获取切片
|
|
|
|
Args:
|
|
project_id: 项目ID
|
|
step: 处理步骤
|
|
|
|
Returns:
|
|
切片列表
|
|
"""
|
|
return self.find_by(project_id=project_id, processing_step=step) |