Trying to fill a cell with pandas with None and/or none but return a string

I'm trying to use Python 3.11 and pandas 2.0.1 to automate a process in work that needs to be a .xlsx file, but I'm facing some issues. I developed the following function:

def add_nan_rows(df, column):
    new_rows = []
    previous_value = None

    for _, row in df.iterrows():
        current_value = row[column]

        if previous_value is not None and current_value != previous_value:
            x = []
            for i in range(0, len(df.columns)):
                x.append(None)
            new_row = pd.DataFrame([pd.Series(x)], columns=df.columns)
            new_rows.append(dict(new_row))

        new_rows.append(dict(row))
        previous_value = current_value
        
    new_df = pd.DataFrame(new_rows, columns = df.columns)
    return new_df

It isn't the most elegant but kinda works in exception that in place of the nothing we spected with None-type variables, but it returned to me a string.

I also tried to substitute None with np.nan or np.NaN and all returned the following string:

0   NaN
Name: NUM_DOCUMENTO, dtype: float64

I also tried to not use a list, but put it directly in the dataframe, but didn't matter.

Can someone explain to me what I am doing wrong, please?

IIUC, you can fix/adjust your custom func this way :

def add_nan_rows(df, column):
    new_rows = []
    previous_value = None

    for _, row in df.iterrows():
        current_value = row[column]

        if previous_value is not None and current_value != previous_value:
            new_rows.append(pd.Series([""] * len(df.columns), index=df.columns, name=""))

        new_rows.append(row)
        previous_value = current_value
    
    new_df = pd.DataFrame(new_rows, columns=df.columns)

    return new_df

Another variant with groupby and concat :

def add_nan_rows(df, column):
    new_df = (
        df.groupby(column, sort=False, group_keys=False)
            .apply(lambda g: pd.concat([g, pd.DataFrame("", columns=df.columns, index=[""])]))
            .iloc[:-1]
    )
    return new_df

Test/Output :

out = add_nan_rows(df, "DUMMY_COL")
​
print(out)

  NUM_DOCUMENTO DUMMY_COL
0             1       foo
1             1       foo
                         
2             2       bar
                         
3             2       baz
4             3       baz
                         
5             3       qux

Excel view :

Input used :

df = pd.DataFrame({
    "NUM_DOCUMENTO": [1, 1, 2, 2, 3, 3],
    "DUMMY_COL": ["foo", "foo", "bar", "baz", "baz", "qux"]}
)