"""ORM models. One table for now: the per-user cached feed (the last completed search)."""
from __future__ import annotations

from datetime import datetime

from sqlalchemy import JSON, Boolean, DateTime, Float, Integer, String, func
from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column


class Base(DeclarativeBase):
    pass


class UserFeed(Base):
    """The user's last completed search — replayed on load so matches survive navigation.
    Upserted on each `search_complete`; read by `get_feed`. One row per user (PK = user_id)."""
    __tablename__ = "user_feed"

    user_id: Mapped[str] = mapped_column(String, primary_key=True)
    prefs: Mapped[dict] = mapped_column(JSON, default=dict)            # the ScoutPrefs that produced it
    opportunities: Mapped[list] = mapped_column(JSON, default=list)    # the curated deck (with match blocks)
    sources: Mapped[dict] = mapped_column(JSON, default=dict)          # per-board counts
    engine: Mapped[str] = mapped_column(String, default="")           # "opus" | "fallback:…"
    scanned: Mapped[int] = mapped_column(Integer, default=0)
    # PER-BOARD search cursors — {board: LAST page fetched}. Next page = last + 1, so a NEW query's
    # first search reads page 1 and stores 1. Resets to {} when the query signature changes (or >24h).
    # Dedup is purely per-user (the seen-net below) — never cross-user, never opaque actor-side state.
    query_sig: Mapped[str] = mapped_column(String, default="")
    cursors: Mapped[dict] = mapped_column(JSON, default=dict)
    search_count: Mapped[int] = mapped_column(Integer, default=0, server_default="0")  # total searches run (engagement)
    # Salary band = AVERAGE of each deck's HIGHEST disclosed salary (₹L). Each deck contributes one
    # peak; the band is the running mean of those peaks. Survives decks that disclose nothing.
    salary_peak_sum: Mapped[float] = mapped_column(Float, default=0.0, server_default="0")
    salary_peak_count: Mapped[int] = mapped_column(Integer, default=0, server_default="0")
    updated_at: Mapped[datetime] = mapped_column(
        DateTime(timezone=True), server_default=func.now(), onupdate=func.now()
    )


class OpportunityState(Base):
    """Per-(user, opportunity) state that SURVIVES deck refreshes. Persists: `applied`/`dismissed`
    status, the resume-builder document ids already generated (never re-pay to re-tailor), and the
    PROGRESSIVE activity flags that feed the dashboard funnel + engagement rank. A job can be viewed
    AND saved AND applied at once, so viewed/saved are booleans (not the single-valued status)."""
    __tablename__ = "opportunity_state"

    user_id: Mapped[str] = mapped_column(String, primary_key=True)
    opportunity_id: Mapped[str] = mapped_column(String, primary_key=True)
    status: Mapped[str | None] = mapped_column(String, nullable=True)              # 'applied' | 'dismissed' | None
    tailored_resume_id: Mapped[str | None] = mapped_column(String, nullable=True)  # resume-builder resume id
    tailored_version_id: Mapped[str | None] = mapped_column(String, nullable=True) # the tailored version
    cover_letter_id: Mapped[str | None] = mapped_column(String, nullable=True)     # the generated cover letter
    seen: Mapped[bool] = mapped_column(Boolean, default=False, server_default="false")    # shown in a deck → exclude from new runs
    viewed: Mapped[bool] = mapped_column(Boolean, default=False, server_default="false")  # opened the job → funnel "Viewed"
    saved: Mapped[bool] = mapped_column(Boolean, default=False, server_default="false")   # swiped right → funnel "Shortlisted"
    updated_at: Mapped[datetime] = mapped_column(
        DateTime(timezone=True), server_default=func.now(), onupdate=func.now()
    )