Creating range of numbers as a column of dataframe

My DataFrame is:

import pandas as pd

df = pd.DataFrame(
        {
           'a': [20, 100],
           'b': [2, 3],
           'dir': ['long', 'short']
        }
 )

Expected output: Creating column x:

     a  b    dir    x
0   20  2   long    [22, 24, 26]
1  100  3  short    [97, 94, 91]

Steps:

x is a list with length of 3. a is starting point of x and b is step that a increases/decreases depending on dir. If df.dir == long x ascends otherwise it descends.

My Attempt based on this answer:

df['x'] = np.arange(0, 3) * df.b + df.a

Which does not produce the expected output.

>Solution :

One possible option :

N = 3

sign = {"long": +1, "short": -1}

df["x"] = [
    np.arange(a+sb, a+sb*(N+1), step=sb)# .tolist() ?
    for a,sb in zip(df["a"], df["b"].mul(df["dir"].map(sign)))
]

Output :

print(df)

     a  b    dir             x
0   20  2   long  [22, 24, 26]
1  100  3  short  [97, 94, 91]

[2 rows x 4 columns]