Unique User Activity Metrics

MediumPython

Problem

Given a collection of records capturing activities, return a mapping where keys are activity types and values are the counts of unique users, and calculate the average time spent on the website per user in seconds.

Asked in the Technical Phone Screen stage. Python coding question on Google Doc with edge cases to discuss.

Function

Implement a function that takes a list of records and returns a tuple: (activity_counts, average_time_per_user).

Each record is a dictionary with keys: user_id (hashable), activity_type (string), and time_spent (non-negative integer seconds).

Output

activity_counts: a dictionary mapping each activity_type to the number of distinct user_id values seen for that activity
average_time_per_user: the total time_spent across all records divided by the number of distinct users

If there are no records, return an empty mapping and 0.0.

Constraints

0 <= len(records) <= 10^5
Each record contains keys: user_id, activity_type, time_spent
user_id is hashable
activity_type is a string
0 <= time_spent <= 10^6

Function Signature