Files
OCR/docengine/app/models/document.py
2026-06-01 21:49:53 +05:30

245 lines
8.8 KiB
Python

from __future__ import annotations
import uuid
from datetime import datetime
from sqlalchemy import BigInteger, Boolean, DateTime, Float, ForeignKey, Integer, String, Text, func
from sqlalchemy.dialects.postgresql import JSONB, UUID
from sqlalchemy.orm import Mapped, mapped_column, relationship
from app.core.database import Base
from app.models.base import TimestampMixin, UUIDPrimaryKeyMixin
class Document(Base, UUIDPrimaryKeyMixin, TimestampMixin):
"""Uploaded document record."""
__tablename__ = "documents"
filename: Mapped[str] = mapped_column(String(500), nullable=False)
original_filename: Mapped[str] = mapped_column(String(500), nullable=False)
content_type: Mapped[str] = mapped_column(String(100), nullable=False)
file_size: Mapped[int] = mapped_column(BigInteger, nullable=False)
checksum: Mapped[str] = mapped_column(String(128), nullable=False, index=True)
storage_path: Mapped[str] = mapped_column(String(1024), nullable=False)
status: Mapped[str] = mapped_column(
String(50),
default="pending",
nullable=False,
index=True,
)
page_count: Mapped[int | None] = mapped_column(Integer, nullable=True)
is_scanned: Mapped[bool | None] = mapped_column(Boolean, nullable=True)
document_metadata: Mapped[dict | None] = mapped_column(JSONB, nullable=True)
error_message: Mapped[str | None] = mapped_column(Text, nullable=True)
uploaded_by: Mapped[uuid.UUID | None] = mapped_column(
UUID(as_uuid=True),
ForeignKey("users.id", ondelete="SET NULL"),
nullable=True,
index=True,
)
pages: Mapped[list[DocumentPage]] = relationship(
"DocumentPage",
back_populates="document",
cascade="all, delete-orphan",
order_by="DocumentPage.page_number",
lazy="selectin",
)
template_matches: Mapped[list[TemplateMatch]] = relationship(
"TemplateMatch",
back_populates="document",
cascade="all, delete-orphan",
lazy="dynamic",
)
def __repr__(self) -> str:
return f"<Document(id={self.id}, filename={self.original_filename}, status={self.status})>"
class DocumentPage(Base, UUIDPrimaryKeyMixin):
"""Individual page within a document."""
__tablename__ = "document_pages"
document_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("documents.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
page_number: Mapped[int] = mapped_column(Integer, nullable=False)
width: Mapped[float] = mapped_column(Float, nullable=False)
height: Mapped[float] = mapped_column(Float, nullable=False)
image_path: Mapped[str | None] = mapped_column(String(1024), nullable=True)
text_content: Mapped[str | None] = mapped_column(Text, nullable=True)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True),
default=func.now(),
server_default=func.now(),
nullable=False,
)
document: Mapped[Document] = relationship("Document", back_populates="pages")
text_blocks: Mapped[list[DocumentTextBlock]] = relationship(
"DocumentTextBlock",
back_populates="page",
cascade="all, delete-orphan",
order_by="DocumentTextBlock.sequence",
lazy="selectin",
)
images: Mapped[list[DocumentImage]] = relationship(
"DocumentImage",
back_populates="page",
cascade="all, delete-orphan",
lazy="selectin",
)
tables: Mapped[list[DocumentTable]] = relationship(
"DocumentTable",
back_populates="page",
cascade="all, delete-orphan",
lazy="selectin",
)
def __repr__(self) -> str:
return f"<DocumentPage(id={self.id}, document_id={self.document_id}, page={self.page_number})>"
class DocumentTextBlock(Base, UUIDPrimaryKeyMixin):
"""Extracted text block from a document page."""
__tablename__ = "document_text_blocks"
page_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("document_pages.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
text: Mapped[str] = mapped_column(Text, nullable=False)
x: Mapped[float] = mapped_column(Float, nullable=False)
y: Mapped[float] = mapped_column(Float, nullable=False)
width: Mapped[float] = mapped_column(Float, nullable=False)
height: Mapped[float] = mapped_column(Float, nullable=False)
confidence: Mapped[float | None] = mapped_column(Float, nullable=True)
font_family: Mapped[str | None] = mapped_column(String(255), nullable=True)
font_size: Mapped[float | None] = mapped_column(Float, nullable=True)
font_color: Mapped[str | None] = mapped_column(String(50), nullable=True)
font_style: Mapped[str | None] = mapped_column(String(50), nullable=True)
block_type: Mapped[str] = mapped_column(
String(50),
default="text",
nullable=False,
)
sequence: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True),
default=func.now(),
server_default=func.now(),
nullable=False,
)
page: Mapped[DocumentPage] = relationship("DocumentPage", back_populates="text_blocks")
def __repr__(self) -> str:
return f"<DocumentTextBlock(id={self.id}, type={self.block_type}, text={self.text[:50]})>"
class DocumentImage(Base, UUIDPrimaryKeyMixin):
"""Extracted image from a document page."""
__tablename__ = "document_images"
page_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("document_pages.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
x: Mapped[float] = mapped_column(Float, nullable=False)
y: Mapped[float] = mapped_column(Float, nullable=False)
width: Mapped[float] = mapped_column(Float, nullable=False)
height: Mapped[float] = mapped_column(Float, nullable=False)
image_path: Mapped[str] = mapped_column(String(1024), nullable=False)
image_type: Mapped[str] = mapped_column(
String(50),
default="figure",
nullable=False,
)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True),
default=func.now(),
server_default=func.now(),
nullable=False,
)
page: Mapped[DocumentPage] = relationship("DocumentPage", back_populates="images")
def __repr__(self) -> str:
return f"<DocumentImage(id={self.id}, type={self.image_type})>"
class DocumentTable(Base, UUIDPrimaryKeyMixin):
"""Extracted table from a document page."""
__tablename__ = "document_tables"
page_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("document_pages.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
x: Mapped[float] = mapped_column(Float, nullable=False)
y: Mapped[float] = mapped_column(Float, nullable=False)
width: Mapped[float] = mapped_column(Float, nullable=False)
height: Mapped[float] = mapped_column(Float, nullable=False)
rows: Mapped[int] = mapped_column(Integer, nullable=False)
columns: Mapped[int] = mapped_column(Integer, nullable=False)
data: Mapped[dict | None] = mapped_column(JSONB, nullable=True)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True),
default=func.now(),
server_default=func.now(),
nullable=False,
)
page: Mapped[DocumentPage] = relationship("DocumentPage", back_populates="tables")
def __repr__(self) -> str:
return f"<DocumentTable(id={self.id}, rows={self.rows}, cols={self.columns})>"
class TemplateMatch(Base, UUIDPrimaryKeyMixin):
"""Template matching result for a document."""
__tablename__ = "template_matches"
document_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("documents.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
format_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True),
ForeignKey("document_formats.id", ondelete="CASCADE"),
nullable=False,
index=True,
)
confidence_score: Mapped[float] = mapped_column(Float, nullable=False)
match_details: Mapped[dict | None] = mapped_column(JSONB, nullable=True)
selected: Mapped[bool] = mapped_column(Boolean, default=False, nullable=False)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True),
default=func.now(),
server_default=func.now(),
nullable=False,
)
document: Mapped[Document] = relationship("Document", back_populates="template_matches")
template: Mapped[DocumentFormat] = relationship("DocumentFormat")
def __repr__(self) -> str:
return f"<TemplateMatch(id={self.id}, doc={self.document_id}, score={self.confidence_score})>"