from datasets import load_dataset
import streamlit as st


@st.cache()
def get_repo_names():
    list_of_repo_names = ["lvwerra/test", "lvwerra/test2", "numpy/numpy"]
    return list_of_repo_names

st.title("Am I in The Stack?")
st.markdown("This tool lets you check if a repository under a given username is part of [The Stack dataset](https://huggingface.co/datasets/bigcode/the-stack).")
repo_names = get_repo_names()

username = st.text_input("GitHub Username:")

if st.button("Check!"):
    list_of_repos = [repo_name for repo_name in repo_names if repo_name.split("/")[0]==username]
    
    if len(list_of_repos)==0:
        st.markdown("There is **no repository** under that username in The Stack.")
    else:
        if len(list_of_repos)==1:
            st.markdown("There is **1 repository** under that username in The Stack:")
        else:
            st.markdown(f"There are **{len(list_of_repos)} repositories** under that username in The Stack:")
        st.text("\n".join(list_of_repos))