import datetime
import json
import os

import gradio as gr
import huggingface_hub as hfh
from apscheduler.schedulers.background import BackgroundScheduler


DATASET_ID = "albertvillanova/datasets-report"
DATASET_PATH = "dataset"
DATA_DIR = "data"
DATA_PATH = f"{DATASET_PATH}{DATA_DIR}"


def pull_dataset_repo(repo_id=DATASET_ID, repo_path=DATASET_PATH):
    token = os.environ.get('HUB_TOKEN')
    repo = hfh.Repository(
        local_dir=repo_path,
        clone_from=repo_id,
        repo_type="dataset",
        use_auth_token=token,
    )
    repo.git_pull()
    return repo


def greet(name):
    return "Hello " + name + "!!"


# iface = gr.Interface(fn=greet, inputs="text", outputs="text")
# iface.launch()


def update_datasets():
    # Retrieve datasets
    datasets = hfh.list_datasets()
    # Save dataset IDs
    repo = pull_dataset_repo()
    os.makedirs(DATA_PATH, exist_ok=True)
    today = datetime.datetime.now(datetime.timezone.utc).date().isoformat()
    with repo.commit(f"Add {today} data file"):
        with open(f"data/{today}.json", "w") as f:
            json.dump([ds.id for ds in sorted(datasets, key=lambda item: item.id)], f)


scheduler = BackgroundScheduler()
scheduler.add_job(update_datasets, trigger="cron", hour=0, minute=1, timezone=datetime.timezone.utc)
scheduler.start()


demo = gr.Interface(fn=greet, inputs="text", outputs="text")
demo.launch()