import pandas as pd
import plotly.express as px


def build_bias_distribution_chart(summary: dict):
    rows = []

    for source, stats in summary.items():
        biased = stats.get("Biased", 0)
        not_biased = stats.get("Not Biased", stats.get("Not_Biased", 0))
        total = stats.get("total", biased + not_biased)

        rows.append(
            {
                "Source": source,
                "Biased": biased,
                "Not biased": not_biased,
                "Total": total,
            }
        )

    df = pd.DataFrame(rows)

    if df.empty:
        return None

    df = df.sort_values("Total", ascending=False)
    df_melted = df.melt(
        id_vars=["Source", "Total"],
        value_vars=["Biased", "Not biased"],
        var_name="Classification",
        value_name="Articles",
    )

    fig = px.bar(
        df_melted,
        x="Source",
        y="Articles",
        color="Classification",
        barmode="group",
        text="Articles",
        color_discrete_map={
            "Biased": "#c24138",
            "Not biased": "#247857",
        },
    )

    fig.update_traces(
        textposition="outside",
        marker_line_width=0,
        cliponaxis=False,
    )
    fig.update_layout(
        height=430,
        margin=dict(l=12, r=12, t=24, b=12),
        paper_bgcolor="rgba(0,0,0,0)",
        plot_bgcolor="rgba(0,0,0,0)",
        bargap=0.26,
        legend=dict(
            orientation="h",
            yanchor="bottom",
            y=1.02,
            xanchor="right",
            x=1,
            title=None,
        ),
        xaxis=dict(
            title=None,
            tickangle=-20,
            showgrid=False,
            linecolor="#d8dee9",
        ),
        yaxis=dict(
            title="Articles",
            gridcolor="#e8edf4",
            zeroline=False,
        ),
        font=dict(color="#15202b", family="Arial, sans-serif"),
    )

    return fig

def build_lean_bias_chart(results: list) -> object:
    from collections import defaultdict

    lean_counts = defaultdict(lambda: {"Biased": 0, "Not biased": 0})

    for article in results:
        lean = article.get("source_bias", "Unknown")
        label = article.get("text_label", "Unknown")
        if label == "Biased":
            lean_counts[lean]["Biased"] += 1
        elif label == "Not Biased":
            lean_counts[lean]["Not biased"] += 1

    rows = []
    for lean, counts in lean_counts.items():
        rows.append({
            "Lean": lean,
            "Biased": counts["Biased"],
            "Not biased": counts["Not biased"],
        })

    df = pd.DataFrame(rows)
    if df.empty:
        return None

    lean_order = ["Left", "Center-Left", "Center", "Center-Right", "Right", "Unknown"]
    df["Lean"] = pd.Categorical(df["Lean"], categories=lean_order, ordered=True)
    df = df.sort_values("Lean")

    df_melted = df.melt(
        id_vars="Lean",
        value_vars=["Biased", "Not biased"],
        var_name="Classification",
        value_name="Articles",
    )

    fig = px.bar(
        df_melted,
        x="Lean",
        y="Articles",
        color="Classification",
        barmode="group",
        text="Articles",
        color_discrete_map={"Biased": "#c24138", "Not biased": "#247857"},
    )

    fig.update_traces(textposition="outside", marker_line_width=0, cliponaxis=False)
    fig.update_layout(
        height=380,
        margin=dict(l=12, r=12, t=24, b=12),
        paper_bgcolor="rgba(0,0,0,0)",
        plot_bgcolor="rgba(0,0,0,0)",
        bargap=0.3,
        legend=dict(orientation="h", yanchor="bottom", y=1.02, xanchor="right", x=1, title=None),
        xaxis=dict(title=None, showgrid=False, linecolor="#d8dee9"),
        yaxis=dict(title="Articles", gridcolor="#e8edf4", zeroline=False),
        font=dict(color="#15202b", family="Arial, sans-serif"),
    )

    return fig