"""Dashboard metrics and KPI components."""

import streamlit as st


def render_kpi_row(
    total_nudge: int,
    high_count: int,
    medium_count: int,
    risk_score: float,
    citation_total: int,
) -> None:
    """Render key metrics row with 5 KPIs.

    Args:
        total_nudge: Total negative mentions
        high_count: HIGH tier count
        medium_count: MEDIUM tier count
        risk_score: Calculated risk score
        citation_total: Total citation sources
    """
    kpi1, kpi2, kpi3, kpi4, kpi5 = st.columns(5)

    with kpi1:
        st.metric(
            label="총 부정 언급",
            value=f"{total_nudge}건",
            help="AI가 자사 브랜드를 부정적으로 언급한 답변 수",
        )

    with kpi2:
        st.metric(
            label="🔴 HIGH (즉시 대응)",
            value=f"{high_count}건",
            help="≥85% 확신도 - 즉시 대응 권장",
        )

    with kpi3:
        st.metric(
            label="🟡 MEDIUM (검토)",
            value=f"{medium_count}건",
            help="70-85% 확신도 - 검토 필요",
        )

    with kpi4:
        st.metric(
            label="리스크 점수",
            value=f"{risk_score:.1f}",
            help="HIGH=100%, MEDIUM=50%, LOW=20% 가중 평균",
        )

    with kpi5:
        st.metric(
            label="총 인용 소스",
            value=f"{citation_total}개",
            help="AI 답변에서 인용된 총 소스 수",
        )


def render_verification_stats(
    total_verified: int,
    false_positives_count: int,
    true_negatives_count: int,
) -> None:
    """Render LLM verification statistics row.

    Args:
        total_verified: Total verified items
        false_positives_count: False positive count
        true_negatives_count: True negative count
    """
    stat_col1, stat_col2, stat_col3, stat_col4 = st.columns(4)

    with stat_col1:
        st.metric("검증 완료", f"{total_verified}건")

    with stat_col2:
        fp_rate = (false_positives_count / total_verified * 100) if total_verified > 0 else 0
        st.metric("오탐 (False Positive)", f"{false_positives_count}건", f"{fp_rate:.1f}%")

    with stat_col3:
        tn_rate = (true_negatives_count / total_verified * 100) if total_verified > 0 else 0
        st.metric("진음성 (True Negative)", f"{true_negatives_count}건", f"{tn_rate:.1f}%")

    with stat_col4:
        if total_verified > 0:
            st.metric("오탐률", f"{fp_rate:.1f}%", delta=None)
        else:
            st.metric("오탐률", "N/A")


def render_polarity_stats(
    positive_count: int,
    neutral_count: int,
    negative_count: int,
) -> None:
    """Render polarity distribution statistics.

    Args:
        positive_count: Positive sentiment count
        neutral_count: Neutral sentiment count
        negative_count: Negative sentiment count
    """
    total_answers = positive_count + neutral_count + negative_count

    pol_col1, pol_col2, pol_col3, pol_col4 = st.columns(4)

    with pol_col1:
        st.metric("전체 분석", f"{total_answers:,}건")

    with pol_col2:
        pos_rate = (positive_count / total_answers * 100) if total_answers > 0 else 0
        st.metric("😊 긍정", f"{positive_count:,}건", f"{pos_rate:.1f}%")

    with pol_col3:
        neu_rate = (neutral_count / total_answers * 100) if total_answers > 0 else 0
        st.metric("😐 중립", f"{neutral_count:,}건", f"{neu_rate:.1f}%")

    with pol_col4:
        neg_rate = (negative_count / total_answers * 100) if total_answers > 0 else 0
        st.metric("😞 부정", f"{negative_count:,}건", f"{neg_rate:.1f}%")


def render_feedback_stats(feedback_stats: dict) -> None:
    """Render feedback statistics row.

    Args:
        feedback_stats: Dict with feedback counts and accuracy
    """
    fb_total = feedback_stats.get("total_feedback", 0)
    fb_correct = feedback_stats.get("correct_count", 0)
    fb_wrong = feedback_stats.get("wrong_count", 0)
    fb_ambiguous = feedback_stats.get("ambiguous_count", 0)
    accuracy = feedback_stats.get("accuracy_rate", 0)

    fb_col1, fb_col2, fb_col3, fb_col4, fb_col5 = st.columns(5)

    with fb_col1:
        st.metric(
            label="총 피드백",
            value=f"{fb_total}건",
            help="사용자가 제출한 총 피드백 수",
        )

    with fb_col2:
        st.metric(
            label="👍 정확",
            value=f"{fb_correct}건",
            delta=f"{fb_correct/fb_total*100:.0f}%" if fb_total > 0 else None,
            delta_color="normal",
            help="정확하다고 평가된 분석 수",
        )

    with fb_col3:
        st.metric(
            label="👎 오류",
            value=f"{fb_wrong}건",
            delta=f"{fb_wrong/fb_total*100:.0f}%" if fb_total > 0 else None,
            delta_color="inverse",
            help="틀렸다고 평가된 분석 수",
        )

    with fb_col4:
        st.metric(
            label="🤔 애매",
            value=f"{fb_ambiguous}건",
            help="판단하기 어려운 경우",
        )

    with fb_col5:
        st.metric(
            label="정확도",
            value=f"{accuracy:.1f}%",
            help="정확 / (정확 + 오류) 비율",
        )