triplets.py

triplets.py

"""
Given a log file of user_id, timestamp & page visited find the 10 most common triplets, where a triplet is an occurrence of 3 pages visited sequentially by the same user. 

1, 1, /home
1, 2, /shipments
1, 3, /labels
1, 4  /something
1, 5  /something
2, 4, /home
...


triplet a list of 3 pages visited in order by the same user
"""
log = """1, 1, /home
1, 2, /shipments
1, 3, /labels
1, 4,  /something
1, 5,  /something
2, 4, /home
"""

from typing import Tuple, List

def parse_line(log_line: str) -> Tuple[int, int, str]:
    parsed_line = [i.strip() for i in log_line.split(',')]
    return (int(parsed_line[0]), int(parsed_line[1]), parsed_line[2])

def analyze_log(log_string: str) -> List[str]:
    lines = log_string.splitlines()
    parsed_lines = [parse_line(line) for line in lines]
    
    
    pages_by_user = {} # {user: [page]}
    triplets_count = {} #{(page, page, page): count}
    for line in parsed_lines:
        user = line[0]
        # timestamp = line[1]
        page = line[2]
        
        if user in pages_by_user:
            pages_by_user[user].append(page)
            if len(pages_by_user[user]) >= 3:
                key = tuple(pages_by_user[user][-3:])
                triplets_count[key] = 1 if key not in triplets_count else triplets_count[key] + 1
        else:
            pages_by_user[user] = [page]
    
    triplets_by_count = {} #{count: List[(page, page, page)]}
    for triplet, count in triplets_count.items():
        triplets_by_count[count] = [triplet] if count not in triplets_by_count else triplets_by_count[count] + [triplet]
    
    result = []
    for i in range(max(triplets_by_count),-1,-1):
        if i in triplets_by_count:
            result = result + triplets_by_count[i]
        if len(result) >= 10:
            break

    return result[:10]

print(analyze_log(log))
        
            
    
            
    
    



def say_hello():
    print('Hello, World')

for i in range(5):
    say_hello()