algorithm.py

import pandas as pd


def build_dataframe(elements: list, drop=True) -> pd.DataFrame:
    frames = []
    for element in elements:
        frames.append(pd.DataFrame(element).dropna())

    df = pd.concat(frames)
    df.reset_index(inplace=True, drop=True)
    if drop:
        df = df.loc[df.Type != "OL-Drop-In"]
    return df


def group_names(df: pd.DataFrame) -> pd.DataFrame:
    lectures = []
    days = ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"]

    for day in days:
        i = df.loc[df.Day == day]
        lectures.append(
            i.groupby(["Day", "Start", "End", "Unit"], as_index=False).agg(
                {"Names": " ".join}
            )
        )
    lectures = pd.concat(lectures).reset_index(drop=True)

    del df["Names"]
    df = df.drop_duplicates(["Day", "Start", "End", "Group", "Building"])

    df.Day = pd.Categorical(df.Day, categories=days, ordered=True)
    df = df.sort_values("Day").reset_index(drop=True)

    df = df.merge(lectures)
    return df