1
|
from Utilities import FolderProcessor, ConfigureFunctions
|
2
|
from Utilities.Database import DatabaseLoader
|
3
|
|
4
|
# Path to crawled data
|
5
|
CRAWLED_DATA_PATH = "CrawledData/"
|
6
|
# Path to processed data
|
7
|
PROCESSED_DATA_PATH = "ProcessedData/"
|
8
|
# Path to dataset crawler implementations
|
9
|
CRAWLER_LIB_PATH = "DatasetCrawler."
|
10
|
# Path to dataset processor implementations
|
11
|
PROCESSOR_LIB_PATH = "DatasetProcessing."
|
12
|
|
13
|
|
14
|
def crawl_data(config):
|
15
|
"""
|
16
|
Imports dataset crawler in DatasetCrawler/"dataset_name"Crawler.py
|
17
|
runs crawler.
|
18
|
|
19
|
Args:
|
20
|
config: loaded configuration file of dataset
|
21
|
"""
|
22
|
dataset_name = config["dataset-name"]
|
23
|
|
24
|
my_function = __import__(CRAWLER_LIB_PATH + dataset_name + "Crawler", globals(), locals(), ['crawl']).crawl
|
25
|
my_function(config)
|
26
|
|
27
|
dataset_name += '/'
|
28
|
|
29
|
|
30
|
def process_data(dataset_name):
|
31
|
"""
|
32
|
Goes trough every not processed file(not contained in CrawledData/"dataset_name"/ignore.txt)
|
33
|
Imports dataset processor in DatasetProcessing/"dataset_name"Processor.py
|
34
|
Runs processor on every file
|
35
|
After successful processing updates ignore.txt
|
36
|
|
37
|
Args:
|
38
|
dataset_name: name of dataset that has existing configuration file
|
39
|
"""
|
40
|
dataset_path = dataset_name + '/'
|
41
|
|
42
|
process_file_func = __import__(PROCESSOR_LIB_PATH + dataset_name + "Processor", globals(), locals(),
|
43
|
['process_file']).process_file
|
44
|
|
45
|
not_processed_files = FolderProcessor.list_of_all_files(CRAWLED_DATA_PATH + dataset_path)
|
46
|
|
47
|
for not_processed_file in not_processed_files:
|
48
|
process_file_func(CRAWLED_DATA_PATH + dataset_path + not_processed_file)
|
49
|
FolderProcessor.update_ignore_set(CRAWLED_DATA_PATH + dataset_path, not_processed_file)
|
50
|
|
51
|
|
52
|
def validate_process_data(config):
|
53
|
"""
|
54
|
Function goes through newly processed data and checks theirs status
|
55
|
|
56
|
Args:
|
57
|
config: loaded configuration file of dataset
|
58
|
|
59
|
Returns:
|
60
|
boolean variable TRUE/FALSE.
|
61
|
Data processed correctly - TRUE
|
62
|
Wrong format or NEW unknown devices - FALSE
|
63
|
"""
|
64
|
processed_devices_set = FolderProcessor.get_devices_set(PROCESSED_DATA_PATH + config["dataset-name"] + '/')
|
65
|
unknown_devices_set = FolderProcessor.get_unknown_devices_set(config, processed_devices_set)
|
66
|
unknown_devices_size = len(unknown_devices_set)
|
67
|
|
68
|
if unknown_devices_size != 0:
|
69
|
print("There is " + str(unknown_devices_size) + " unknown devices\n")
|
70
|
print("Adding devices to " + config["dataset-name"] + " config file\n")
|
71
|
ConfigureFunctions.update_configuration(config["dataset-name"], unknown_devices_set)
|
72
|
return False
|
73
|
|
74
|
for device in config["devices"]:
|
75
|
device = config["devices"][device]
|
76
|
if device["x"] == "UNKNOWN!" or device["y"] == "UNKNOWN!":
|
77
|
print(config["dataset-name"] + " Config file contains devices with UNKOWN! values please update them\n")
|
78
|
return False
|
79
|
|
80
|
return True
|
81
|
|
82
|
|
83
|
def load_data_to_database(config):
|
84
|
"""
|
85
|
Goes trough every not loaded file(not contained in ProcessedData/ignore.txt)
|
86
|
loads data appends coordination from configurations
|
87
|
and exports it into the database
|
88
|
After successful exporting updates ignore.txt
|
89
|
|
90
|
Args:
|
91
|
config: loaded configuration file of dataset
|
92
|
"""
|
93
|
dataset_name = config["dataset-name"]
|
94
|
dataset_path = dataset_name + '/'
|
95
|
|
96
|
# get all unprocessed files from dataset
|
97
|
not_loaded_files = FolderProcessor.list_of_all_files(PROCESSED_DATA_PATH + dataset_path)
|
98
|
|
99
|
# load every file
|
100
|
for not_loaded_file in not_loaded_files:
|
101
|
# load processed data
|
102
|
processed_data = DatabaseLoader.get_data_from_file(not_loaded_file, config)
|
103
|
# load processed data to database
|
104
|
DatabaseLoader.load_data_to_database(dataset_name, processed_data)
|
105
|
FolderProcessor.update_ignore_set(PROCESSED_DATA_PATH + dataset_path, not_loaded_file)
|
106
|
|
107
|
|
108
|
def run_full_pipeline(dataset_name):
|
109
|
"""
|
110
|
Loads config file and starts full pipeline
|
111
|
-crawl data
|
112
|
-process data
|
113
|
-load data to database
|
114
|
|
115
|
Args:
|
116
|
dataset_name: name of dataset that has existing configuration file
|
117
|
"""
|
118
|
config = ConfigureFunctions.load_configuration(dataset_name)
|
119
|
crawl_data(config)
|
120
|
process_data(config["dataset-name"])
|
121
|
|
122
|
validation_test = validate_process_data(config)
|
123
|
|
124
|
if validation_test:
|
125
|
load_data_to_database(config)
|
126
|
|