import itertools
from datetime import datetime

import pandas as pd
import plotly.graph_objects as go
from dash import dcc, html

from src.constants import COLORS
from src.utils import normalize_prices


def setup_ticker_selection(initial_tickers: list[str]) -> html.Div:
    ticker_selection = html.Div(
        [
            dcc.Dropdown(
                id="ticker-selection",
                options=[{"label": ticker, "value": ticker} for ticker in initial_tickers],
                value=list(initial_tickers),
                multi=True,
                placeholder="No tickers selected...",
                searchable=False,
                clearable=False,
                persisted_props=[],
                style={
                    "flex": "3",
                    "fontFamily": "'Courier New', Courier, monospace",
                    "fontWeight": "bold",
                },
            ),
            dcc.Input(
                id="ticker-input",
                type="text",
                placeholder="Enter ticker symbol...",
                style={
                    "flex": "1",
                    "padding": "10px",
                    "fontSize": "14px",
                    "fontFamily": "'Courier New', Courier, monospace",
                    "borderRadius": "5px",
                    "border": "1px solid #ccc",
                },
                n_submit=0,
            ),
        ],
        style={
            "display": "flex",
            "gap": "10px",
            "marginTop": "10px",
            "marginLeft": "80px",
            "marginRight": "80px",
        },
    )
    return ticker_selection


def setup_interval_buttons() -> tuple[html.Div, list[str], dict[str, int]]:
    button_style = {
        "padding": "10px 20px",
        "borderRadius": "10px",
        "cursor": "pointer",
        "fontFamily": "'Courier New', Courier, monospace",
        "fontWeight": "bold",
        "textAlign": "center",
        "marginBottom": "5px",
        "marginTop": "5px",
    }
    interval_buttons_html = html.Div(
        [
            html.Button("ytd", id="btn-ytd", n_clicks=0, style=button_style),
            html.Button("1mo", id="btn-1mo", n_clicks=0, style=button_style),
            html.Button("6mo", id="btn-6mo", n_clicks=0, style=button_style),
            html.Button("1y", id="btn-1y", n_clicks=0, style=button_style),
            html.Button("2y", id="btn-2y", n_clicks=0, style=button_style),
            html.Button("3y", id="btn-3y", n_clicks=0, style=button_style),
            html.Button("5y", id="btn-5y", n_clicks=0, style=button_style),
            html.Button("10y", id="btn-10y", n_clicks=0, style=button_style),
        ],
        style={
            "gap": "5px",
            "display": "flex",
            "flexWrap": "wrap",
            "marginLeft": "80px",
            "marginRight": "80px",
        },
    )
    interval_buttons_ids = [
        "btn-ytd",
        "btn-1mo",
        "btn-6mo",
        "btn-1y",
        "btn-2y",
        "btn-3y",
        "btn-5y",
        "btn-10y",
    ]
    interval_offsets = {
        "btn-ytd": max(1, (datetime.now() - datetime(datetime.now().year, 1, 1)).days),
        "btn-1mo": 30,
        "btn-6mo": 182,
        "btn-1y": 365,
        "btn-2y": 2 * 365,
        "btn-3y": 3 * 365,
        "btn-5y": 5 * 365,
        "btn-10y": 10 * 365,
    }
    return interval_buttons_html, interval_buttons_ids, interval_offsets


def plot_prices(
    timestamps: pd.DatetimeIndex,
    prices: pd.DataFrame,
    rolling_changes: pd.DataFrame,
    idx_range: tuple[int, int],
) -> go.Figure:
    idx0, idx1 = idx_range
    date_range = [timestamps[idx0], timestamps[idx1]]
    prices_normalized = normalize_prices(prices, date_range)

    fig = go.Figure()

    # rangeslider plot
    colors = itertools.cycle(COLORS)
    for asset in prices.columns:
        fig.add_trace(
            go.Scatter(
                x=timestamps,
                y=rolling_changes[asset],
                line=dict(color=next(colors)),
                xaxis="x1",
                yaxis="y1",
                showlegend=False,
            )
        )

    # main plot
    colors = itertools.cycle(COLORS)
    for asset in prices_normalized.columns:
        y_values = 100 * prices_normalized[asset]
        formatted_values = [f"{val:+.2f}%" for val in y_values]
        fig.add_trace(
            go.Scatter(
                x=timestamps,
                y=y_values,
                customdata=formatted_values,
                line=dict(width=3, color=next(colors)),
                name=asset,
                xaxis="x2",
                yaxis="y2",
                hovertemplate=(
                    "<b style='font-family:Courier New,monospace'>"
                    "%{customdata} %{fullData.name}"
                    "</b><br>"
                    "<span style='font-family:Courier New,monospace'>%{x}</span>"
                    "<extra></extra>"
                ),
            )
        )

    # dummy traces to show ticks on the right
    for _ in prices_normalized.columns:
        fig.add_trace(
            go.Scatter(
                x=[],
                y=[],
                xaxis="x2",
                yaxis="y3",
                showlegend=False,
            )
        )

    # configure axes
    xaxis1_dict = dict(rangeslider=dict(visible=True, thickness=0.1), tickangle=-30, nticks=20)
    xaxis2_dict = dict(matches="x1", showticklabels=False, nticks=20, showgrid=True)
    xaxis1_dict["range"] = date_range  # type: ignore
    xaxis2_dict["range"] = date_range  # type: ignore
    yaxis1_dict = dict(showticklabels=False)
    yaxis2_dict = dict(
        autorange=True,
        title="relative price change",
        nticks=12,
        tickformat="+d",
        ticksuffix="%",
        ticks="outside",
    )
    yaxis3_dict = dict(
        matches="y2",
        overlaying="y2",
        side="right",
        nticks=12,
        tickformat="+d",
        ticksuffix="%",
        ticks="outside",
    )

    fig.update_layout(
        xaxis1=xaxis1_dict,
        yaxis1=yaxis1_dict,
        xaxis2=xaxis2_dict,
        yaxis2=yaxis2_dict,
        yaxis3=yaxis3_dict,
        uirevision="constant",  # prevent resets from the xrange compression
        font=dict(family="Courier New, Monospace", size=14, weight="bold"),
        legend=dict(
            title=dict(text="Tickers: "),
            orientation="h",
            x=0.0,
            y=1.0,
            xanchor="left",
            yanchor="bottom",
        ),
        margin=dict(t=50, b=10),
        template="plotly",
        height=600,
    )

    return fig