Take rankings from 2 csv files and use them to compute the best pairings.

PeopleMatching.py

from matching import Player, HospitalResident
from collections import defaultdict
import csv

    
def read_csv_ranks_v1(fn, other_fields=2):

    users = {}
    items = []

    with open(fn, newline='') as csvfile:
        
        reader = csv.reader(csvfile)
        rows = [row for row in reader]
        
        header = rows[0]
        cols = len(header)

        for i in range(3, cols - other_fields):
            # Parse "Rank your preferences (#1 is your first preference, #10 is your least) [XXXX]"
            item = header[i].split(") ")[-1][1:-1]
            items.append(item)

        num_items = len(items)

        for row in rows[1:]:
            name = row[2].split(" ")[0]
            ranks = [[None, int(r)] for r in row[3:cols-other_fields]]
            for i in range(num_items):
                ranks[i][0] = items[i]
            ranks.sort(key=lambda rank: rank[1])
            users[name] = [rank[0] for rank in ranks]

    return users


def read_csv_ranks_v2(fn, other_fields=2):

    users = {}
    rank_vals = []

    with open(fn, newline='') as csvfile:
        
        reader = csv.reader(csvfile)
        rows = [row for row in reader]
        
        header = rows[0]
        cols = len(header)

        for i in range(4, cols - other_fields):
            # Parse "Rank your preferences (#1 is your first preference, #10 is your least) [X]"
            rank = header[i].split(") ")[-1][1:-1]
            rank_vals.append(int(rank))

        num_ranks = len(rank_vals)

        for row in rows[1:]:
            name = row[2]
            ranks = [(item, rank_vals[k]) for k, item in enumerate(row[4:cols-other_fields])]
            ranks.sort(key=lambda rank: rank[1])
            users[name] = [rank[0] for rank in ranks]

    return users


def trim_rankings(a_dict, b_dict):
    """Delete values from a that dont exist in b"""
    b_list = list(b_dict)
    for name, prefs in a_dict.items():
        a_prefs = list(prefs)
        for b in a_prefs:
            if b not in b_list:
                prefs.remove(b)


if __name__ == "__main__":

    people = read_csv_ranks_v1('People.csv')
    companies = read_csv_ranks_v2('Company.csv')

    trim_rankings(people, companies)
    trim_rankings(companies, people)

    company_sizes = {
        'name': 1,
        ...
    }

    print(companies.keys())
    print(people.keys())

    residents = [Player(name, prefs)
                 for name, prefs in people.items()]
    
    hospitals = [Player(name, prefs, capacity=company_sizes[name])
                 for name, prefs in companies.items()]

    hr_solver = HospitalResident(suitors=residents, reviewers=hospitals)
    solu = hr_solver.solve()
    
    for comp, names in solu.items():
        print('{} ({}): {}'.format(comp, company_sizes[str(comp)], str(names)))