Distributed-query/input.py at master · Bonis98/Distributed-query · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
import logging

import pandas as pd

from node import Node
from relation import Relation


def read_input(input_path):
    logging.info("Reading data from input files...")
    nodes, global_Ap = read_tree(input_path)
    relations = read_relations(input_path, nodes)
    subjects, avg_comp_price, avg_transfer_price = read_subjects(input_path)
    authorizations = read_authorizations(input_path)
    return nodes[0], relations, subjects, authorizations, avg_comp_price, avg_transfer_price, global_Ap


def read_tree(input_path):
    logging.debug('Reading tree structure...')
    global_Ap = set()
    nodes = list()
    df = pd.read_csv(input_path + 'tree.csv')
    df['parent'] = df['parent'].fillna(value=0)
    df = df.fillna(value='')
    df = df.astype({'ID': 'int', 'parent': 'int'})
    df = df.astype({'Ap': 'str', 'Ae': 'str', 'As': 'str'})
    for idx, row in df.iterrows():
        global_Ap = global_Ap.union(set(row['Ap']))
        multi_attr = False
        if row['operation'] == 'selection':
            if (len(row['Ap']) + len(row['Ae']) + len(row['As'])) > 1:
                multi_attr = True
        if not idx:
            node = Node(
                operation=row['operation'], Ap=row['Ap'], Ae=row['Ae'], As=row['As'],
                print_label=row['print_label'], group_attr=row['group_attr'], select_multi_attr=multi_attr)
        else:
            node = Node(
                operation=row['operation'], Ap=row['Ap'], Ae=row['Ae'], As=row['As'],
                print_label=row['print_label'], group_attr=row['group_attr'],
                select_multi_attr=multi_attr, parent=nodes[row['parent'] - 1])
        nodes.append(node)
    return nodes, global_Ap


def read_relations(input_path, nodes: list):
    logging.debug('Reading relations...')
    relations = list()
    df = pd.read_csv(input_path + 'relations.csv')
    df = df.fillna(value='')
    df = df.astype('str')
    df = df.astype({'node_id': int})
    for idx, row in df.iterrows():
        relation = Relation(
            name=row['name'], storage_provider=row['provider'], primary_key=row['primary_key'],
            plain_attr=row['plain_attr'], enc_attr=row['enc_attr'], attr=row['attr'],
            enc_costs=row['enc_costs'], dec_costs=row['dec_costs'], size=row['size'])
        relations.append(relation)
        nodes[row['node_id'] - 1].relation = relation
    return relations


def read_subjects(input_path):
    logging.debug('Reading subjects...')
    df = pd.read_csv(input_path + 'subjects.csv')
    if df.isnull().values.any():
        raise ValueError("Input: Subjects dataframe can't contain NaN values")
    df = df.set_index('subject')
    df['sum'] = df['comp_price'] + df['transfer_price']
    df = df.sort_values(by=['sum'])
    df = df.drop(columns=['sum'])
    subjects = df.T.to_dict('dict')
    avg_comp_price = df['comp_price'].median()
    avg_transfer_price = df['transfer_price'].median()
    return subjects, int(avg_comp_price), int(avg_transfer_price)


def read_authorizations(input_path):
    logging.debug('Reading authorizations...')
    df = pd.read_csv(input_path + 'authorizations.csv')
    df = df.fillna(value='')
    return df.set_index('subject').T.to_dict('dict')