import pandas as pd

# Define the dataframes
municip = {
    "muni_id": [1401, 1402, 1407, 1415, 1419, 1480, 1480, 1427, 1484],
    "muni_name": ["Har", "Par", "Ock", "Ste", "Tjo", "Gbg", "Gbg", "Sot", "Lys"],
    "new_muni_id": [1401, 1402, 1480, 1415, 1415, 1480, 1480, 1484, 1484],
    "new_muni_name": ["Har", "Par", "Gbg", "Ste", "Ste", "Gbg", "Gbg", "Lys", "Lys"],
    "new_node_id": ["HAR1", "PAR1", "GBG2", "STE1", "STE1", "GBG1", "GBG2", "LYS1", "LYS1"]
}
df_1 = pd.DataFrame(municip)

demand = {
    "period": [1, 2, 3, 4, 5],
    "1401": [2, 4, 4, 1, 2],
    "1402": [1, 1, 3, 3, 5],
    "1407": [2, 4, 4, 1, 2],
    "1415": [1, 1, 3, 3, 5],
    "1419": [1, 1, 3, 3, 5],
    "1480": [1, 1, 3, 3, 5],
    "1427": [2, 4, 4, 1, 2],
    "1484": [1, 2, 3, 4, 5]
}
df_2 = pd.DataFrame(demand)

def profiling(df_1, df_2):
    # Merge demand and municipality data
    merged_df = pd.merge(df_2, df_1, left_on="period", right_on="muni_id")
    
    # Calculate demand by number of nodes
    merged_df["demand"] = merged_df["1401"] / (merged_df["new_muni_id"] == "HAR1").sum() + \
                         merged_df["1402"] / (merged_df["new_muni_id"] == "PAR1").sum() + \
                         merged_df["1480"] / (merged_df["new_muni_id"] == "GBG2").sum() + \
                         merged_df["1427"] / (merged_df["new_muni_id"] == "LYS1").sum()
    
    # Pivot table to get demand by node and period
    final_df = merged_df.pivot_table(index="period", columns="new_node_id", values="demand", aggfunc="sum")
    
    return final_df

final_df = profiling(df_1, df_2)
print(final_df)