"""Author RAG Chatbot SaaS — Custom Q&A Model.

Allows authors to define custom question-answer pairs that override
the RAG pipeline retrieval for specific queries.
"""

from sqlalchemy import Boolean, Float, ForeignKey, Integer, String, Text
from sqlalchemy.orm import Mapped, mapped_column, relationship

from app.models.base import Base, TimestampMixin, generate_uuid


class CustomQA(Base, TimestampMixin):
    """Custom Q&A pair — checked before RAG pipeline."""

    __tablename__ = "custom_qa"

    id: Mapped[str] = mapped_column(String(36), primary_key=True, default=generate_uuid)
    author_id: Mapped[str] = mapped_column(
        String(36), ForeignKey("users.id", ondelete="CASCADE"), nullable=False, index=True
    )
    book_id: Mapped[str | None] = mapped_column(
        String(36), ForeignKey("books.id", ondelete="SET NULL"), nullable=True, index=True
    )
    question: Mapped[str] = mapped_column(String(500), nullable=False)
    answer: Mapped[str] = mapped_column(Text, nullable=False)
    priority: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
    is_active: Mapped[bool] = mapped_column(Boolean, default=True, nullable=False)
    match_count: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
    match_threshold: Mapped[float] = mapped_column(Float, default=0.85, nullable=False)
    category: Mapped[str | None] = mapped_column(String(50), nullable=True)

    # Phase 3B: Stores JSON-serialized list[float] for semantic cosine similarity.
    # NULL = embedding not yet generated → Jaccard fallback used until re-saved.
    embedding_json: Mapped[str | None] = mapped_column(Text, nullable=True)

    # Relationships
    author: Mapped["User"] = relationship("User")
    book: Mapped["Book"] = relationship("Book")

    def __repr__(self) -> str:
        return f"<CustomQA id={self.id} q={self.question[:30]}>"