Small-Multiple Distributions With displot

Calder Health has clinics whose waiting patterns are genuinely different, and one chart with every clinic overlaid is a mess of colour. The alternative is a row of small charts, one per clinic, all on shared axes so shapes can be compared without untangling them. That layout stops being readable past a handful, so the build first reports how many small charts the data would produce.

Task: Print how many small charts the data would produce.

Input

The first line holds one integer n, the number of readings. Each of the next n lines holds a clinic name and one reading, separated by a single space. Clinic names contain no spaces.

Output

One line holding a single whole number: how many different clinic names appear. One small chart is produced per clinic, however many readings that clinic has.

Example:

Input:
4
A 1
B 2
A 3
C 4

Output:
3