posts has columns post_id, post_date (datetime) and keywords. views has columns post_id and viewer_id. A post is spam when its keywords string contains "spam".

For each day on which at least one post was viewed, compute the percentage of views that landed on a spam post. Return a DataFrame with columns day (the date as a "YYYY-MM-DD" string) and spam_pct, rounded to 2 decimal places, sorted by day with the index renumbered from 0.

Posts nobody viewed do not appear, since a day with no views has no percentage.

Input

posts =
   post_id  post_date        keywords
0        1 2022-04-01  sale spam deal
1        2 2022-04-01         holiday
2        3 2022-04-02            spam
views =
   post_id  viewer_id
0        1          9
1        1          8
2        2          7
3        3          6

Output

          day  spam_pct
0  2022-04-01     66.67
1  2022-04-02    100.00