__all__ = ['block', 'make_clickable_model', 'make_clickable_user', 'get_submissions']

import gradio as gr
import pandas as pd
import re
import pandas as pd
import os
import json

from src.about import *

global data_component, filter_component


def get_baseline_df():
    df = pd.read_csv(CSV_RESULT_PATH)
    present_columns = ["Method"] + checkbox_group.value
    df = df[present_columns]
    return df

def update_yaml(representation_name, benchmark_type, human_file_path, skempi_file_path):
    with open("./src/bin/probe_config.yaml", 'r') as file:
        yaml_data = yaml.safe_load(file)

    yaml_data['representation_name'] = representation_name
    yaml_data['benchmark'] = benchmark_type
    yaml_data['representation_file_human'] = human_file
    yaml_data['representation_file_affinity'] = skempi_file

    with open("./src/bin/probe_config.yaml", "w") as file:
        yaml.dump(yaml_data, file)

    return None

def add_new_eval(
    human_file,
    skempi_file,
    model_name_textbox: str,
    revision_name_textbox: str,
    benchmark_type: str,
):
    representation_name = model_name_textbox if revision_name_textbox == '' else revision_name_textbox

    update_yaml(representation_name, benchmark_type, human_file, skempi_file)
    
    # Save human and skempi files under ./src/data/representation_vectors using pandas
    print(human_file)
    df = pd.read_csv(human_file)
    print(df.head().to_string())
    return None
    if human_file is not None:
        human_df = pd.read_csv(human_file)
        human_df.to_csv(f"./src/data/representation_vectors/{representation_name}_human.csv", index=False)

    return None

block = gr.Blocks()

with block:
    gr.Markdown(
        LEADERBOARD_INTRODUCTION
    )
    with gr.Tabs(elem_classes="tab-buttons") as tabs:
        # table jmmmu bench
        with gr.TabItem("🏅 PROBE Benchmark", elem_id="probe-benchmark-tab-table", id=1):
            # selection for column part:
            checkbox_group = gr.CheckboxGroup(
                choices=TASK_INFO,
                label="Benchmark Type",
                interactive=True,
            ) # user can select the evaluation dimension

            baseline_value = get_baseline_df()
            baseline_header = ["Method"] + checkbox_group.value
            baseline_datatype = ['markdown'] + ['number'] * len(checkbox_group.value)

            data_component = gr.components.Dataframe(
                value=baseline_value,
                headers=baseline_header,
                type="pandas",
                datatype=baseline_datatype,
                interactive=False,
                visible=True,
                )

        # table 5
        with gr.TabItem("📝 About", elem_id="probe-benchmark-tab-table", id=2):
            with gr.Row():
                gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")

        with gr.TabItem("🚀 Submit here! ", elem_id="probe-benchmark-tab-table", id=3):
            with gr.Row():
                gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")

            with gr.Row():
                gr.Markdown("# ✉️✨ Submit your model's representation files here!", elem_classes="markdown-text")

            with gr.Row():
                with gr.Column():
                    model_name_textbox = gr.Textbox(
                        label="Model name",
                        )
                    revision_name_textbox = gr.Textbox(
                        label="Revision Model Name",
                    )
                    # Selection for benchmark type from (similartiy, family, function, affinity) to eval the representations (chekbox)
                    benchmark_type = gr.CheckboxGroup(
                        choices=TASK_INFO,
                        label="Benchmark Type",
                        interactive=True,
                    )

            with gr.Column():
                human_file = gr.components.File(label="Click to Upload the representation file (csv) for Human dataset", file_count="single", type='filepath')
                skempi_file = gr.components.File(label="Click to Upload the representation file (csv) for SKEMPI dataset", file_count="single", type='filepath')
    
                submit_button = gr.Button("Submit Eval")
                submission_result = gr.Markdown()
                submit_button.click(
                    add_new_eval,
                    inputs = [
                        human_file,
                        skempi_file,
                        model_name_textbox,
                        revision_name_textbox,
                        benchmark_type
                    ],
                )

    def refresh_data():
        value = get_baseline_df()

        return value

    with gr.Row():
        data_run = gr.Button("Refresh")
        data_run.click(
            refresh_data, outputs=[data_component]
        )

    with gr.Accordion("Citation", open=False):
        citation_button = gr.Textbox(
            value=CITATION_BUTTON_TEXT,
            label=CITATION_BUTTON_LABEL,
            elem_id="citation-button",
            show_copy_button=True,
        )

block.launch()