What is the best way to slice a dataframe up to the first instance of a mask?

This is my DataFrame:

import pandas as pd
df = pd.DataFrame(
    {
        'a': [10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70],
        'b': [1, 1, 1, -1, -1, -2, -1, 2, 2, -2, -2, 1, -2],
    }
)

The mask is:

mask = (
    (df.b == -2) &
    (df.b.shift(1) > 0)
)

Expected output: slicing df up to the first instance of the mask:

   a  b
0  10  1
1  15  1
2  20  1
3  25 -1
4  30 -1
5  35 -2
6  40 -1
7  45  2
8  50  2

The first instance of the mask is at row 9. So I want to slice the df up to this index.

This is what I have tried. It works but I am not sure if it is the best way:

idx = df.loc[mask.cumsum().eq(1) & mask].index[0]
result = df.iloc[:idx]

>Solution :

You can filter by inverted mask with Series.cummax:

out = df[~mask.cummax()]
print (out)

    a  b
0  10  1
1  15  1
2  20  1
3  25 -1
4  30 -1
5  35 -2
6  40 -1
7  45  2
8  50  2

How it working:

print (df.assign(mask=mask,
                 cumax=mask.cummax(),
                 inv_cummax=~mask.cummax()))

     a  b   mask  cumax  inv_cummax
0   10  1  False  False        True
1   15  1  False  False        True
2   20  1  False  False        True
3   25 -1  False  False        True
4   30 -1  False  False        True
5   35 -2  False  False        True
6   40 -1  False  False        True
7   45  2  False  False        True
8   50  2  False  False        True
9   55 -2   True   True       False
10  60 -2  False   True       False
11  65  1  False   True       False
12  70 -2   True   True       False