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

# Load the CSV data
df = pd.read_csv("iobench.csv")  # Replace with the actual file path

# Initialize Dash app
app = Dash(__name__)

# Layout
app.layout = html.Div(
    [
        html.H1("IOBench Results Viewer", style={"textAlign": "center"}),

        # Filters
        html.Div(
            [
                html.Label("Filter by Label:"),
                dcc.Dropdown(
                    id="label-filter",
                    options=[{"label": label, "value": label} for label in df["label"].unique()],
                    value=df["label"].unique().tolist(),
                    multi=True,
                ),
                html.Label("Filter by Test Name:"),
                dcc.Dropdown(
                    id="test-filter",
                    options=[{"label": test, "value": test} for test in df["test_name"].unique()],
                    value=df["test_name"].unique().tolist(),
                    multi=True,
                ),
            ],
            style={"width": "25%", "display": "inline-block", "verticalAlign": "top", "padding": "10px"},
        ),

        # Graphs
        html.Div(
            [
                dcc.Graph(id="throughput-graph"),
                dcc.Graph(id="latency-graph"),
            ],
            style={"width": "70%", "display": "inline-block", "padding": "10px"},
        ),
    ]
)

# Callbacks
@app.callback(
    [Output("throughput-graph", "figure"), Output("latency-graph", "figure")],
    [Input("label-filter", "value"), Input("test-filter", "value")],
)
def update_graphs(selected_labels, selected_tests):
    # Filter data
    filtered_df = df[df["label"].isin(selected_labels) & df["test_name"].isin(selected_tests)]

    # Throughput Graph
    throughput_fig = go.Figure()
    for label in filtered_df["label"].unique():
        subset = filtered_df[filtered_df["label"] == label]
        throughput_fig.add_trace(
            go.Bar(
                x=subset["test_name"],
                y=subset["iops"],
                name=f"{label} - IOPS",
            )
        )
        throughput_fig.add_trace(
            go.Bar(
                x=subset["test_name"],
                y=subset["bandwidth_kibps"],
                name=f"{label} - Bandwidth (KiB/s)",
            )
        )
    throughput_fig.update_layout(
        title="Throughput (IOPS and Bandwidth)",
        xaxis_title="Test Name",
        yaxis_title="Value",
        barmode="group",
    )

    # Latency Graph
    latency_fig = go.Figure()
    for label in filtered_df["label"].unique():
        subset = filtered_df[filtered_df["label"] == label]
        latency_fig.add_trace(
            go.Scatter(
                x=subset["test_name"],
                y=subset["latency_mean_ms"],
                mode="markers+lines",
                name=f"{label} - Latency Mean (ms)",
                error_y=dict(
                    type="data",
                    array=subset["latency_stddev_ms"],
                    visible=True,
                ),
            )
        )
    latency_fig.update_layout(
        title="Latency with Standard Deviation",
        xaxis_title="Test Name",
        yaxis_title="Latency (ms)",
    )

    return throughput_fig, latency_fig


if __name__ == "__main__":
    app.run_server(debug=True)