pandas: Select from a Series by label and position

Model loc and iloc on a Series: one selects by label, the other by position.

What to build

A program that answers loc and iloc lookups on a Series.

Requirements

  • The Series is given as LABEL VALUE lines. Each query is loc LABEL or iloc POSITION.
  • Print the value, or KeyError: 'LABEL' for an unknown label, or IndexError: single positional indexer is out-of-bounds for a position outside -n to n - 1. As in Python and pandas, a negative position counts from the end: -1 is the last value.

Input

Line 1: n. Then n lines LABEL VALUE. Next line: q. Then q queries.

Output

q lines.

Constraints

1 ≤ n, q ≤ 100; labels are distinct words; values are whole numbers.

Example

Input:

3
apple 3
bread 2
milk 1
4
loc bread
iloc 0
loc tea
iloc 3

Output:

2
3
KeyError: 'tea'
IndexError: single positional indexer is out-of-bounds

Why it works

loc looks the label up in the index and iloc counts positions, from 0 at the start or from -1 at the end, which is why the same Series can answer both kinds of question.

Examples

Example 1 · Input
3
apple 3
bread 2
milk 1
4
loc bread
iloc 0
loc tea
iloc 3
Expected output
2
3
KeyError: 'tea'
IndexError: single positional indexer is out-of-bounds
Example 2 · Input
2
x 10
y 20
2
iloc 1
loc x
Expected output
20
10

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

Keep the labels and values in two lists, or in a dictionary plus a list of labels.

Hint 2

For iloc, a position p is valid when -n <= p < n, and Python's own list indexing then does the rest.

Solution

Show one correct solution (try it yourself first)
n = int(input())
labels = []
values = []
for _ in range(n):
    label, value = input().split()
    labels.append(label)
    values.append(int(value))
q = int(input())
for _ in range(q):
    kind, key = input().split()
    if kind == "loc":
        print(values[labels.index(key)] if key in labels else f"KeyError: '{key}'")
    else:
        p = int(key)
        print(values[p] if -n <= p < n else "IndexError: single positional indexer is out-of-bounds")

Lesson: Series: one labelled column →