app.py

import gradio as gr
import json
import pandas as pd
from urllib.request import urlopen
from urllib.error import URLError
import re
from datetime import datetime

# Constants
CITATION_BUTTON_TEXT = r"""@misc{2023opencompass,
    title={OpenCompass: A Universal Evaluation Platform for Foundation Models},
    author={OpenCompass Contributors},
    howpublished = {\url{https://github.com/open-compass/opencompass}},
    year={2023}
}"""
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
# 开发环境
# DATA_URL_BASE = "http://opencompass.oss-cn-shanghai.aliyuncs.com/dev-assets/research-rank/research-data.REALTIME."
# DATA_URL_BASE = "./s1test"
# 生产环境
DATA_URL_BASE = "http://opencompass.oss-cn-shanghai.aliyuncs.com/assets/research-rank/research-data.REALTIME."


def find_latest_data_url():
    """Find the latest available data URL by trying different dates."""
    today = datetime.now()
    for i in range(365):
        date = today.replace(day=today.day - i)
        date_str = date.strftime("%Y%m%d")
        url = f"{DATA_URL_BASE}{date_str}.json"
        try:
            urlopen(url)
            return url, date_str
        except URLError:
            continue
    breakpoint()
    return None, None


def get_latest_data():
    """Get latest data URL and update time"""
    data_url, update_time = find_latest_data_url()
    if not data_url:
        raise Exception("Could not find valid data URL")
    formatted_update_time = datetime.strptime(update_time, "%Y%m%d").strftime("%Y-%m-%d")
    return data_url, formatted_update_time


def get_leaderboard_title(update_time):
    return f"# Supported Datasets List (Last Updated: {update_time})"


MAIN_DESCRIPTION = """## The List of Datasets Supported by OpenCompass
Testing line.
- All configurations and datsets can be found in [**OpenCompass**: A Toolkit for Evaluation of LLMs](https://github.com/open-compass/opencompass)🏆.
"""




def load_data(data_url):
    response = urlopen(data_url)

    with open('s1.json','r',encoding='utf8') as f:
        data = json.load(f)

    return data


def build_main_table(data):
    df = pd.DataFrame(data).transpose()
    columns = {
        'name': 'Name', 'category': 'Category', 'article': 'Article Address',
    }
    df = df[list(columns.keys())].rename(columns=columns)
    return df


DATA_CATEGORY = ['med', 'law', 'code']

def filter_table1(df, data_category):
    filtered_df = df.copy()
    if data_category:
        mask = pd.Series(False, index=filtered_df.index)
        for category in data_category:
            mask |= filtered_df['Category'] == category
        filtered_df = filtered_df[mask]

    return filtered_df


def calculate_column_widths(df):
    column_widths = []
    for column in df.columns:
        header_length = len(str(column))
        max_content_length = df[column].astype(str).map(len).max()
        width = max(header_length * 10, max_content_length * 8) + 20
        width = max(160, min(400, width))
        column_widths.append(width)
    return column_widths


class DataState:
    def __init__(self):
        self.current_df = None


data_state = DataState()


def create_interface():
    empty_df = pd.DataFrame(columns=[
        'Name', 'Category', 'Article Address'
    ])

    def load_initial_data():
        try:
            data_url, update_time = get_latest_data()
            data = load_data(data_url)
            new_df = build_main_table(data)
            data_state.current_df = new_df
            filtered_df = filter_table1(new_df, DATA_CATEGORY)
            return get_leaderboard_title(update_time), filtered_df.sort_values("Name", ascending=True)
        except Exception as e:
            print(f"Error loading initial data: {e}")
            return "# Supported Datasets List (Error loading data)", empty_df

    def refresh_data():
        try:
            data_url, update_time = get_latest_data()
            data = load_data(data_url)
            new_df = build_main_table(data)
            data_state.current_df = new_df
            filtered_df = filter_table1(new_df, DATA_CATEGORY)
            return get_leaderboard_title(update_time), filtered_df.sort_values("Name", ascending=True)
        except Exception as e:
            print(f"Error refreshing data: {e}")
            return None, None

    def update_table(category):
        if data_state.current_df is None:
            return empty_df
        filtered_df = filter_table1(data_state.current_df, category)
        return filtered_df.sort_values("Name", ascending=True)

    initial_title, initial_data = load_initial_data()

    with gr.Blocks() as demo:
        title_comp = gr.Markdown(initial_title)

        with gr.Tabs() as tabs:
            with gr.TabItem("Dataset List", elem_id='main'):
                gr.Markdown(MAIN_DESCRIPTION)

                with gr.Row():
                    with gr.Column():
                        category_filter = gr.CheckboxGroup(
                            choices=DATA_CATEGORY,
                            value=DATA_CATEGORY,
                            label='Category',
                            interactive=True,
                        )

                with gr.Column():
                    table = gr.DataFrame(
                        value=initial_data,
                        interactive=False,
                        wrap=False,
                        column_widths=calculate_column_widths(initial_data),
                    )

                refresh_button = gr.Button("Refresh Data")

                def refresh_and_update():
                    title, data = refresh_data()
                    return title, data

                refresh_button.click(
                    fn=refresh_and_update,
                    outputs=[title_comp, table],
                )

                category_filter.change(
                    fn=update_table,
                    inputs=[category_filter],
                    outputs=table,
                )


        with gr.Row():
            with gr.Accordion("Citation", open=False):
                citation_button = gr.Textbox(
                    value=CITATION_BUTTON_TEXT,
                    label=CITATION_BUTTON_LABEL,
                    elem_id='citation-button',
                    lines=6,  # 增加行数
                    max_lines=8,  # 设置最大行数
                    show_copy_button=True  # 添加复制按钮使其更方便使用
                )

    return demo


if __name__ == '__main__':
    demo = create_interface()
    demo.queue()
    demo.launch(server_name='0.0.0.0')