import streamlit as st
import pandas as pd

from shared_tools_module import csv_url_to_dataframe
from pasbdc_data_cleaning import clean_center_name, remove_duplicate_client_records, remove_api_testing_clients
from constants_module import NEOSERRA_COLUMNS, OUT_COLUMNS

@st.cache_data
def cached_csv_url_to_dataframe(export_module_url) -> pd.DataFrame:
    return csv_url_to_dataframe(export_module_url)

@st.cache_data
def get_df_centers(export_module_url:str) -> pd.DataFrame:
    raw_client_df = csv_url_to_dataframe(export_module_url).copy()

    clean_center_name(raw_client_df)
    raw_client_df = remove_duplicate_client_records(raw_client_df)
    raw_client_df = remove_api_testing_clients(raw_client_df)

    return raw_client_df[NEOSERRA_COLUMNS.center].unique()