modules.py

import pandas as pd

SUPPORTED_TYPES = [".csv", ".json", ".xlsx"]

def hello_world(): return "hello world!"

def load_file(file):
    """
    Takes a file given by Streamlit and loads into a DataFrame.
    Returns a DataFrame, metadata, and result string.

    @param file: File uploaded into streamlit.
    @rtype: tuple
    @return: A tuple of format (pd.DataFrame, (str, str), str).
    """
    df = None

    if file is None: return df, ("", ""), ""

    filename = file.name
    extension = filename.split(".")[-1] 
    metadata = (filename, extension)

    import_functions = {
        "csv": pd.read_csv,
        "json": pd.read_json,
        "xlsx": pd.read_excel
    }
    try:
        reader = import_functions.get(extension, None)
        if reader is None: 
            return df, metadata, f"Error: Invalid extension '{extension}'"
        df = reader(file)
        rows, columns = df.shape
        return df, metadata, f"File '{filename}' loaded successfully.\nFound {rows} rows, {columns} columns."
    except Exception as error:
        return df, metadata, f"Error: Unable to read file '{filename}' ({type(error)}: {error})"

def data_cleaner(df, drop_missing=False, remove_duplicates=True):
    """
    Takes a DataFrame and removes empty and duplicate entries.

    @type df: pd.DataFrame
    @param df: A DataFrame of uncleaned data.
    @type drop_missing: bool
    @param drop_missing: Determines if rows with any missing values are dropped ("any"), or just empty rows ("all").
    @type remove_duplicates: bool
    @param remove_duplicates: Determines if duplicate rows are removed.
    @rtype: pd.DataFrame
    @return: A DataFrame with requested cleaning applied
    """
    df = df.dropna(how="any" if drop_missing else "all")
    if remove_duplicates: df = df.drop_duplicates()
    return df

def data_anonymizer(df):
    return df