Who Worked on What

A small studio logs work as a list of timesheet entries. Each entry is a tuple (person, project, hours); the same person and project can appear in many entries.

Write timesheet(entries) that returns a dictionary with one key per person. The value is a tuple (total_hours, main_project):

• total_hours is everything that person logged, over all projects;
• main_project is the project on which that person logged the most hours in total; if several projects tie, take the one that comes first alphabetically.

Return an empty dictionary when there are no entries. The helper work_log(n, seed) returns n random entries.

Examples

Input:  entries = [("ana", "web", 3), ("ben", "app", 2.5), ("ana", "app", 1.5),
                   ("ana", "web", 0.5), ("ben", "web", 2.5)]
Output: {"ana": (5.0, "web"), "ben": (5.0, "app")}
Explanation: ana has 3.5 hours on web and 1.5 on app. ben has 2.5 on each,
a tie, and "app" comes before "web".

Input:  entries = [("jo", "docs", 2)]
Output: {"jo": (2, "docs")}

Constraints

• 0 <= len(entries) <= 10**5
• hours are positive multiples of 0.25 and at most 8
• names and projects are non-empty lowercase strings

Goals

• Group values with a dictionary of dictionaries
• Walk a dictionary with .items() and break ties deliberately