first

backend/app/models/ocr.py

@@ -0,0 +1,122 @@
+"""
+Tool_OCR - OCR Models
+Database models for OCR batches, files, and results
+"""
+
+from sqlalchemy import Column, Integer, String, DateTime, Float, Text, ForeignKey, Enum, JSON
+from sqlalchemy.orm import relationship
+from datetime import datetime
+import enum
+
+from app.core.database import Base
+
+
+class BatchStatus(str, enum.Enum):
+    """Batch processing status"""
+    PENDING = "pending"
+    PROCESSING = "processing"
+    COMPLETED = "completed"
+    PARTIAL = "partial"  # Some files failed
+    FAILED = "failed"
+
+
+class FileStatus(str, enum.Enum):
+    """Individual file processing status"""
+    PENDING = "pending"
+    PROCESSING = "processing"
+    COMPLETED = "completed"
+    FAILED = "failed"
+
+
+class OCRBatch(Base):
+    """OCR batch processing tracking"""
+
+    __tablename__ = "paddle_ocr_batches"
+
+    id = Column(Integer, primary_key=True, index=True)
+    user_id = Column(Integer, ForeignKey("paddle_ocr_users.id", ondelete="CASCADE"), nullable=False, index=True)
+    batch_name = Column(String(255), nullable=True)
+    status = Column(Enum(BatchStatus), default=BatchStatus.PENDING, nullable=False, index=True)
+    total_files = Column(Integer, default=0, nullable=False)
+    completed_files = Column(Integer, default=0, nullable=False)
+    failed_files = Column(Integer, default=0, nullable=False)
+    created_at = Column(DateTime, default=datetime.utcnow, nullable=False, index=True)
+    started_at = Column(DateTime, nullable=True)
+    completed_at = Column(DateTime, nullable=True)
+
+    # Relationships
+    user = relationship("User", back_populates="ocr_batches")
+    files = relationship("OCRFile", back_populates="batch", cascade="all, delete-orphan")
+
+    @property
+    def progress_percentage(self) -> float:
+        """Calculate progress percentage"""
+        if self.total_files == 0:
+            return 0.0
+        return (self.completed_files / self.total_files) * 100
+
+    def __repr__(self):
+        return f"<OCRBatch(id={self.id}, status='{self.status}', progress={self.progress_percentage:.1f}%)>"
+
+
+class OCRFile(Base):
+    """Individual file in an OCR batch"""
+
+    __tablename__ = "paddle_ocr_files"
+
+    id = Column(Integer, primary_key=True, index=True)
+    batch_id = Column(Integer, ForeignKey("paddle_ocr_batches.id", ondelete="CASCADE"), nullable=False, index=True)
+    filename = Column(String(255), nullable=False)
+    original_filename = Column(String(255), nullable=False)
+    file_path = Column(String(512), nullable=False)
+    file_size = Column(Integer, nullable=False)  # Size in bytes
+    file_format = Column(String(20), nullable=False)  # png, jpg, pdf, etc.
+    status = Column(Enum(FileStatus), default=FileStatus.PENDING, nullable=False, index=True)
+    error_message = Column(Text, nullable=True)
+    retry_count = Column(Integer, default=0, nullable=False)  # Number of retry attempts
+    created_at = Column(DateTime, default=datetime.utcnow, nullable=False)
+    started_at = Column(DateTime, nullable=True)
+    completed_at = Column(DateTime, nullable=True)
+    processing_time = Column(Float, nullable=True)  # Processing time in seconds
+
+    # Relationships
+    batch = relationship("OCRBatch", back_populates="files")
+    result = relationship("OCRResult", back_populates="file", uselist=False, cascade="all, delete-orphan")
+
+    def __repr__(self):
+        return f"<OCRFile(id={self.id}, filename='{self.filename}', status='{self.status}')>"
+
+
+class OCRResult(Base):
+    """OCR processing result with structure and images"""
+
+    __tablename__ = "paddle_ocr_results"
+
+    id = Column(Integer, primary_key=True, index=True)
+    file_id = Column(Integer, ForeignKey("paddle_ocr_files.id", ondelete="CASCADE"), unique=True, nullable=False, index=True)
+
+    # Output file paths
+    markdown_path = Column(String(512), nullable=True)  # Path to Markdown file
+    json_path = Column(String(512), nullable=True)  # Path to JSON file
+    images_dir = Column(String(512), nullable=True)  # Directory containing extracted images
+
+    # OCR metadata
+    detected_language = Column(String(20), nullable=True)  # ch, en, japan, korean
+    total_text_regions = Column(Integer, default=0, nullable=False)
+    average_confidence = Column(Float, nullable=True)
+
+    # Layout structure data (stored as JSON)
+    # Contains: layout elements (title, paragraph, table, image, formula), reading order, bounding boxes
+    layout_data = Column(JSON, nullable=True)
+
+    # Extracted images metadata (stored as JSON)
+    # Contains: list of {image_path, bbox, element_type}
+    images_metadata = Column(JSON, nullable=True)
+
+    created_at = Column(DateTime, default=datetime.utcnow, nullable=False)
+
+    # Relationships
+    file = relationship("OCRFile", back_populates="result")
+
+    def __repr__(self):
+        return f"<OCRResult(id={self.id}, file_id={self.file_id}, language='{self.detected_language}')>"