dada.py

import json
import collections
import csv

f = open('contacts.json')
c = json.load(f)

# c = [
#     {'Id': '0', 'Email': 'Wick', 'Phone': '', 'OrderId': '', 'Contacts': '3'},
#     {'Id': '1', 'Email': 'John', 'Phone': '', 'OrderId': '12345678', 'Contacts': '5'},
#     {'Id': '8', 'Email': 'Chan', 'Phone': '782212345', 'OrderId': '', 'Contacts': '4'},
#     {'Id': '34567', 'Email': '', 'Phone': '682212345', 'OrderId': '12345678', 'Contacts': '2'},
#     {'Id': '78999', 'Email': 'Wick', 'Phone': '682212345', 'OrderId': '', 'Contacts': '4'},
#     {'Id': '80000', 'Email': 'Dada', 'Phone': '', 'OrderId': '', 'Contacts': '3'},
#     {'Id': '80001', 'Email': 'Chan', 'Phone': '', 'OrderId': '87654321', 'Contacts': '5'},
#     {'Id': '80002', 'Email': '', 'Phone': '782212345', 'OrderId': '87654321', 'Contacts': '2'},
# ]

# ans = {}

# for cc in c:
#     cc['Key'] = 

ans = collections.OrderedDict()

for ii in range(500):
    anstmp = collections.OrderedDict()
    print('Hi' + str(ii))
    for cc in c[ii*1000:(ii+1)*1000]:
        find = False
        for aa in anstmp:
            if cc['Email'] in anstmp[aa]['Data'] or cc['Phone'] in anstmp[aa]['Data'] or cc['OrderId'] in anstmp[aa]['Data']:
                if cc['Email'] != '':
                    anstmp[aa]['Data'].add(cc['Email'])
                if cc['Phone'] != '':
                    anstmp[aa]['Data'].add(cc['Phone'])
                if cc['OrderId'] != '':
                    anstmp[aa]['Data'].add(cc['OrderId'])
                anstmp[aa]['Contacts'] += int(cc['Contacts'])
                anstmp[aa]['Ids'].append(str(cc['Id']))
                find = True
                break
        if not find:
            anstmp[int(cc['Id'])] = {'Data': set(), 'Contacts': int(cc['Contacts']), 'Ids': [str(cc['Id'])]}
            if cc['Email'] != '':
                anstmp[int(cc['Id'])]['Data'].add(cc['Email'])
            if cc['Phone'] != '':
                anstmp[int(cc['Id'])]['Data'].add(cc['Phone'])
            if cc['OrderId'] != '':
                anstmp[int(cc['Id'])]['Data'].add(cc['OrderId'])
    ans.update(anstmp)

print('Hihi')
# print(ans[0])

ans2 = collections.OrderedDict()

for aa in ans:
    find = False
    print('Hoho'+str(aa))
    for a2 in ans2:
        if not ans[aa]['Data'].isdisjoint(ans2[a2]['Data']):
            ans2[a2]['Data'].union(ans[aa]['Data'])
            ans2[a2]['Contacts'] += ans[aa]['Contacts']
            ans2[a2]['Ids'].extend(ans[aa]['Ids'])
            ans2[a2]['Ids'] = sorted(ans2[a2]['Ids'])
            find = True
            break
    if not find:
        ans2[aa] = ans[aa]

print('Hohoho')
# ans2 = sorted(ans2)
# print(ans2)

real_ans = collections.OrderedDict()

for a in ans2:
    for i in ans2[a]['Ids']:
        real_ans[int(i)] = {'ticket_trace': '-'.join(sorted(ans2[a]['Ids'])), 'contact': ans2[a]['Contacts']}

# print(sorted(real_ans))
# print(real_ans)

anss = []

for r in sorted(real_ans):
    anss.append([r, real_ans[r]['ticket_trace'] + ', ' + str(real_ans[r]['contact'])])

fields = ['ticket_id', 'ticket_trace/contact']

with open('ans2.csv', 'w') as f2:
    write = csv.writer(f2)
    write.writerow(fields)
    write.writerows(anss)