backend/app/models/dataset.py

from __future__ import annotations

from datetime import datetime, timezone

from sqlalchemy import DateTime, Enum as SAEnum, Integer, String, Text, func
from sqlalchemy.orm import Mapped, mapped_column

from .base import Base, DatasetType


class Dataset(Base):
    """A batch-loaded collection of dogs/users (test fixtures or real imports).

    Existing tables carry a nullable ``dataset_id`` back to here so a whole batch can be listed,
    inspected, and purged as a unit. Cases/matches are linked transitively through the dogs/users
    and are removed by the transactional purge endpoint (see api/datasets.py).
    """

    __tablename__ = "datasets"

    id: Mapped[int] = mapped_column(primary_key=True)
    name: Mapped[str] = mapped_column(String(160))
    type: Mapped[DatasetType] = mapped_column(SAEnum(DatasetType, native_enum=False))
    description: Mapped[str | None] = mapped_column(Text, nullable=True)
    source_path: Mapped[str | None] = mapped_column(String(512), nullable=True)
    dog_count: Mapped[int] = mapped_column(Integer, default=0)
    creation_time: Mapped[datetime] = mapped_column(
        DateTime(timezone=True),
        default=lambda: datetime.now(timezone.utc),
        server_default=func.now(),
    )