327 lines
12 KiB
Python
327 lines
12 KiB
Python
#!/usr/bin/env python3
|
||
# -*- coding: utf-8 -*-
|
||
"""
|
||
翻譯任務資料模型
|
||
|
||
Author: PANJIT IT Team
|
||
Created: 2024-01-28
|
||
Modified: 2024-01-28
|
||
"""
|
||
|
||
import json
|
||
import uuid
|
||
from datetime import datetime, timedelta
|
||
from sqlalchemy.sql import func
|
||
from sqlalchemy import event
|
||
from app import db
|
||
from app.utils.timezone import format_taiwan_time
|
||
|
||
|
||
class TranslationJob(db.Model):
|
||
"""翻譯任務表 (dt_translation_jobs)"""
|
||
__tablename__ = 'dt_translation_jobs'
|
||
|
||
id = db.Column(db.Integer, primary_key=True, autoincrement=True)
|
||
job_uuid = db.Column(db.String(36), unique=True, nullable=False, index=True, comment='任務唯一識別碼')
|
||
user_id = db.Column(db.Integer, db.ForeignKey('dt_users.id'), nullable=False, comment='使用者ID')
|
||
original_filename = db.Column(db.String(500), nullable=False, comment='原始檔名')
|
||
file_extension = db.Column(db.String(10), nullable=False, comment='檔案副檔名')
|
||
file_size = db.Column(db.BigInteger, nullable=False, comment='檔案大小(bytes)')
|
||
file_path = db.Column(db.String(1000), nullable=False, comment='檔案路徑')
|
||
source_language = db.Column(db.String(50), default=None, comment='來源語言')
|
||
target_languages = db.Column(db.JSON, nullable=False, comment='目標語言陣列')
|
||
status = db.Column(
|
||
db.Enum('PENDING', 'PROCESSING', 'COMPLETED', 'FAILED', 'RETRY', name='job_status'),
|
||
default='PENDING',
|
||
comment='任務狀態'
|
||
)
|
||
progress = db.Column(db.Numeric(5, 2), default=0.00, comment='處理進度(%)')
|
||
retry_count = db.Column(db.Integer, default=0, comment='重試次數')
|
||
error_message = db.Column(db.Text, comment='錯誤訊息')
|
||
total_tokens = db.Column(db.Integer, default=0, comment='總token數')
|
||
total_cost = db.Column(db.Numeric(10, 4), default=0.0000, comment='總成本')
|
||
conversation_id = db.Column(db.String(100), comment='Dify對話ID，用於維持翻譯上下文')
|
||
processing_started_at = db.Column(db.DateTime, comment='開始處理時間')
|
||
completed_at = db.Column(db.DateTime, comment='完成時間')
|
||
created_at = db.Column(db.DateTime, default=func.now(), comment='建立時間')
|
||
updated_at = db.Column(
|
||
db.DateTime,
|
||
default=func.now(),
|
||
onupdate=func.now(),
|
||
comment='更新時間'
|
||
)
|
||
deleted_at = db.Column(db.DateTime, comment='軟刪除時間')
|
||
|
||
# 關聯關係
|
||
files = db.relationship('JobFile', backref='job', lazy='dynamic', cascade='all, delete-orphan')
|
||
api_usage_stats = db.relationship('APIUsageStats', backref='job', lazy='dynamic')
|
||
|
||
def __repr__(self):
|
||
return f'<TranslationJob {self.job_uuid}>'
|
||
|
||
def __init__(self, **kwargs):
|
||
"""初始化，自動生成 UUID"""
|
||
super().__init__(**kwargs)
|
||
if not self.job_uuid:
|
||
self.job_uuid = str(uuid.uuid4())
|
||
|
||
def to_dict(self, include_files=False):
|
||
"""轉換為字典格式"""
|
||
data = {
|
||
'id': self.id,
|
||
'job_uuid': self.job_uuid,
|
||
'user_id': self.user_id,
|
||
'original_filename': self.original_filename,
|
||
'file_extension': self.file_extension,
|
||
'file_size': self.file_size,
|
||
'file_path': self.file_path,
|
||
'source_language': self.source_language,
|
||
'target_languages': self.target_languages,
|
||
'status': self.status,
|
||
'progress': float(self.progress) if self.progress else 0.0,
|
||
'retry_count': self.retry_count,
|
||
'error_message': self.error_message,
|
||
'total_tokens': self.total_tokens,
|
||
'total_cost': float(self.total_cost) if self.total_cost else 0.0,
|
||
'conversation_id': self.conversation_id,
|
||
'processing_started_at': format_taiwan_time(self.processing_started_at, "%Y-%m-%d %H:%M:%S") if self.processing_started_at else None,
|
||
'completed_at': format_taiwan_time(self.completed_at, "%Y-%m-%d %H:%M:%S") if self.completed_at else None,
|
||
'created_at': format_taiwan_time(self.created_at, "%Y-%m-%d %H:%M:%S") if self.created_at else None,
|
||
'updated_at': format_taiwan_time(self.updated_at, "%Y-%m-%d %H:%M:%S") if self.updated_at else None,
|
||
'deleted_at': format_taiwan_time(self.deleted_at, "%Y-%m-%d %H:%M:%S") if self.deleted_at else None
|
||
}
|
||
|
||
if include_files:
|
||
data['files'] = [f.to_dict() for f in self.files]
|
||
|
||
return data
|
||
|
||
def update_status(self, status, error_message=None, progress=None):
|
||
"""更新任務狀態"""
|
||
self.status = status
|
||
|
||
if error_message:
|
||
self.error_message = error_message
|
||
|
||
if progress is not None:
|
||
self.progress = progress
|
||
|
||
if status == 'PROCESSING' and not self.processing_started_at:
|
||
self.processing_started_at = datetime.utcnow()
|
||
elif status == 'COMPLETED':
|
||
self.completed_at = datetime.utcnow()
|
||
self.progress = 100.00
|
||
|
||
self.updated_at = datetime.utcnow()
|
||
db.session.commit()
|
||
|
||
def add_original_file(self, filename, file_path, file_size):
|
||
"""新增原始檔案記錄"""
|
||
from pathlib import Path
|
||
stored_name = Path(file_path).name
|
||
|
||
original_file = JobFile(
|
||
job_id=self.id,
|
||
file_type='source',
|
||
original_filename=filename,
|
||
stored_filename=stored_name,
|
||
file_path=file_path,
|
||
file_size=file_size,
|
||
mime_type=self._get_mime_type(filename)
|
||
)
|
||
db.session.add(original_file)
|
||
db.session.commit()
|
||
return original_file
|
||
|
||
def add_translated_file(self, language_code, filename, file_path, file_size):
|
||
"""新增翻譯檔案記錄"""
|
||
from pathlib import Path
|
||
stored_name = Path(file_path).name
|
||
|
||
translated_file = JobFile(
|
||
job_id=self.id,
|
||
file_type='translated',
|
||
language_code=language_code,
|
||
original_filename=filename,
|
||
stored_filename=stored_name,
|
||
file_path=file_path,
|
||
file_size=file_size,
|
||
mime_type=self._get_mime_type(filename)
|
||
)
|
||
db.session.add(translated_file)
|
||
db.session.commit()
|
||
return translated_file
|
||
|
||
def _get_mime_type(self, filename):
|
||
"""取得MIME類型"""
|
||
import mimetypes
|
||
from pathlib import Path
|
||
|
||
ext = Path(filename).suffix.lower()
|
||
mime_map = {
|
||
'.docx': 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
|
||
'.pdf': 'application/pdf',
|
||
'.pptx': 'application/vnd.openxmlformats-officedocument.presentationml.presentation',
|
||
'.xlsx': 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
|
||
'.txt': 'text/plain'
|
||
}
|
||
return mime_map.get(ext, mimetypes.guess_type(filename)[0] or 'application/octet-stream')
|
||
|
||
def get_translated_files(self):
|
||
"""取得翻譯檔案"""
|
||
return self.files.filter_by(file_type='translated').all()
|
||
|
||
def get_original_file(self):
|
||
"""取得原始檔案"""
|
||
return self.files.filter_by(file_type='source').first()
|
||
|
||
def can_retry(self):
|
||
"""是否可以重試"""
|
||
return self.status in ['FAILED', 'RETRY'] and self.retry_count < 3
|
||
|
||
def increment_retry(self):
|
||
"""增加重試次數"""
|
||
self.retry_count += 1
|
||
self.updated_at = datetime.utcnow()
|
||
db.session.commit()
|
||
|
||
def soft_delete(self):
|
||
"""軟刪除任務（保留資料供報表使用）"""
|
||
self.deleted_at = datetime.utcnow()
|
||
self.updated_at = datetime.utcnow()
|
||
db.session.commit()
|
||
|
||
def restore(self):
|
||
"""恢復已刪除的任務"""
|
||
self.deleted_at = None
|
||
self.updated_at = datetime.utcnow()
|
||
db.session.commit()
|
||
|
||
def is_deleted(self):
|
||
"""檢查任務是否已被軟刪除"""
|
||
return self.deleted_at is not None
|
||
|
||
@classmethod
|
||
def get_queue_position(cls, job_uuid):
|
||
"""取得任務在佇列中的位置"""
|
||
job = cls.query.filter_by(job_uuid=job_uuid, deleted_at=None).first()
|
||
if not job:
|
||
return None
|
||
|
||
position = cls.query.filter(
|
||
cls.status == 'PENDING',
|
||
cls.deleted_at.is_(None),
|
||
cls.created_at < job.created_at
|
||
).count()
|
||
|
||
return position + 1
|
||
|
||
@classmethod
|
||
def get_pending_jobs(cls):
|
||
"""取得所有等待處理的任務"""
|
||
return cls.query.filter_by(status='PENDING', deleted_at=None).order_by(cls.created_at.asc()).all()
|
||
|
||
@classmethod
|
||
def get_processing_jobs(cls):
|
||
"""取得所有處理中的任務"""
|
||
return cls.query.filter_by(status='PROCESSING', deleted_at=None).all()
|
||
|
||
@classmethod
|
||
def get_user_jobs(cls, user_id, status=None, limit=None, offset=None, include_deleted=False):
|
||
"""取得使用者的任務列表"""
|
||
query = cls.query.filter_by(user_id=user_id)
|
||
|
||
# 預設排除軟刪除的記錄，除非明確要求包含
|
||
if not include_deleted:
|
||
query = query.filter(cls.deleted_at.is_(None))
|
||
|
||
if status and status != 'all':
|
||
query = query.filter_by(status=status.upper())
|
||
|
||
query = query.order_by(cls.created_at.desc())
|
||
|
||
if limit:
|
||
query = query.limit(limit)
|
||
if offset:
|
||
query = query.offset(offset)
|
||
|
||
return query.all()
|
||
|
||
@classmethod
|
||
def get_statistics(cls, user_id=None, start_date=None, end_date=None, include_deleted=True):
|
||
"""取得統計資料（預設包含所有記錄以確保報表完整性）"""
|
||
query = cls.query
|
||
|
||
# 報表統計預設包含已刪除記錄以確保數據完整性
|
||
if not include_deleted:
|
||
query = query.filter(cls.deleted_at.is_(None))
|
||
|
||
if user_id:
|
||
query = query.filter_by(user_id=user_id)
|
||
|
||
if start_date:
|
||
query = query.filter(cls.created_at >= start_date)
|
||
|
||
if end_date:
|
||
query = query.filter(cls.created_at <= end_date)
|
||
|
||
total = query.count()
|
||
completed = query.filter_by(status='COMPLETED').count()
|
||
failed = query.filter_by(status='FAILED').count()
|
||
processing = query.filter_by(status='PROCESSING').count()
|
||
pending = query.filter_by(status='PENDING').count()
|
||
|
||
return {
|
||
'total': total,
|
||
'completed': completed,
|
||
'failed': failed,
|
||
'processing': processing,
|
||
'pending': pending,
|
||
'success_rate': (completed / total * 100) if total > 0 else 0
|
||
}
|
||
|
||
|
||
class JobFile(db.Model):
|
||
"""檔案記錄表 (dt_job_files)"""
|
||
__tablename__ = 'dt_job_files'
|
||
|
||
id = db.Column(db.Integer, primary_key=True, autoincrement=True)
|
||
job_id = db.Column(db.Integer, db.ForeignKey('dt_translation_jobs.id'), nullable=False, comment='任務ID')
|
||
file_type = db.Column(
|
||
db.Enum('source', 'translated', name='file_type'),
|
||
nullable=False,
|
||
comment='檔案類型'
|
||
)
|
||
language_code = db.Column(db.String(50), comment='語言代碼(翻譯檔案)')
|
||
original_filename = db.Column(db.String(255), nullable=False, comment='原始檔名')
|
||
stored_filename = db.Column(db.String(255), nullable=False, comment='儲存檔名')
|
||
file_path = db.Column(db.String(500), nullable=False, comment='檔案路徑')
|
||
file_size = db.Column(db.BigInteger, default=0, comment='檔案大小')
|
||
mime_type = db.Column(db.String(100), comment='MIME 類型')
|
||
created_at = db.Column(db.DateTime, default=func.now(), comment='建立時間')
|
||
|
||
def __repr__(self):
|
||
return f'<JobFile {self.original_filename}>'
|
||
|
||
def to_dict(self):
|
||
"""轉換為字典格式"""
|
||
return {
|
||
'id': self.id,
|
||
'job_id': self.job_id,
|
||
'file_type': self.file_type,
|
||
'language_code': self.language_code,
|
||
'original_filename': self.original_filename,
|
||
'stored_filename': self.stored_filename,
|
||
'file_path': self.file_path,
|
||
'file_size': self.file_size,
|
||
'mime_type': self.mime_type,
|
||
'created_at': format_taiwan_time(self.created_at, "%Y-%m-%d %H:%M:%S") if self.created_at else None
|
||
}
|
||
|
||
|
||
# 事件監聽器：自動生成 UUID
|
||
@event.listens_for(TranslationJob, 'before_insert')
|
||
def receive_before_insert(mapper, connection, target):
|
||
"""在插入前自動生成 UUID"""
|
||
if not target.job_uuid:
|
||
target.job_uuid = str(uuid.uuid4()) |