-
Notifications
You must be signed in to change notification settings - Fork 0
/
graph_crawler.py
251 lines (224 loc) · 9.38 KB
/
graph_crawler.py
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
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
from waybackcrawler import TemporalPageCrawler, URLFilter
import csv
import sys
import os
import urlparse
import threading
import Queue
import time
import process_manager
# This is a Example Use Case of the Package
# The Script takes a list of nodes as a input it
# Then generates a list of edges for each URL on the specific timestamps
# This is collected for all graphs
class GraphURLFilter(TemporalPageCrawler.URLFilter):
"""
We need to filter and select only those URLS
which are provided in our set. We domain hashing for our task
Basically given a url, we extract the netloc and hash it.
Even though this would allow a significant number of more urls, than exact match
But we are fine with them.
"""
def __init__(self, allowed_urls):
TemporalPageCrawler.URLFilter.__init__(self);
self.allowed_hostnames = self.__create_host_name_hash(allowed_urls);
def __create_host_name_hash(self, allowed_urls):
allowed_hostnames = {};
for url in allowed_urls:
hostname = self.__get_hostname(url);
if hostname != None:
allowed_hostnames[hostname] = True;
return allowed_hostnames;
def __get_hostname(self, url):
hostname = urlparse.urlparse(url).hostname;
if not hostname:
# This could happen because of two issues
# a) Url has our needed link inside
# b) Url does not start with a schema
# Case a)
schema_loc = url.find("http");
if schema_loc != -1:
sub_url = url[schema_loc:];
hostname = urlparse.urlparse(sub_url).hostname;
return hostname;
protocol_loc = url.find('www');
if protocol_loc != -1:
sub_url = "http://" + url[protocol_loc:];
hostname = urlparse.urlparse(sub_url).hostname;
return hostname;
# Case b)
sub_url = "http://" + url;
hostname = urlparse.urlparse(sub_url).hostname;
return hostname
def filter(self, url_list):
filtered_urls = [];
for url in url_list:
hostname = self.__get_hostname(url);
if self.allowed_hostnames.has_key(hostname):
filtered_urls.append(url);
return filtered_urls;
class GraphURLFilterSingleton(GraphURLFilter):
"""
This is a extension of the GRaph url filter which
only allows one url. Hence If the url crawl is completed we remove it
"""
def __init__(self, allowed_urls):
GraphURLFilter.__init__(self);
self.allowed_hostnames = self.__create_host_name_hash(allowed_urls);
def filter(self, url_list):
filtered_urls = [];
for url in url_list:
hostname = self.__get_hostname(url);
if self.allowed_hostnames.has_key(hostname) and \
self.allowed_hostnames[hostname]:
self.allowed_hostnames[hostname] = False
filtered_urls.append(url);
return filtered_urls;
class GraphCrawler:
"""
Grap
"""
def __init__(self, input_file_name, file_delimiter = "\t"):
self.file_delimiter = file_delimiter;
self.input_hash = self.process_csv_file(input_file_name,
file_delimiter);
self.output_file_name = self.__make_output_file_name(
input_file_name);
# Setting up progress bar functionality
self.progress_queue = Queue.Queue();
self.progress_bar_thread = threading.Thread(target = self.progress_bar);
self.stopreuest = threading.Event();
self.refreash_rate = 60;
def __make_output_file_name(self, input_file_name):
fileName, fileExtension = os.path.splitext(input_file_name);
return fileName + ".out";
def process_csv_file(self, in_file_name, delimiter):
result_hash = [];
with open(in_file_name, 'rb') as csv_file:
inp_stream = csv.DictReader(csv_file,
delimiter = delimiter);
for entry in inp_stream:
result_hash.append(entry);
return result_hash;
def progress_bar(self):
print "Starting Collection\n";
message_printed = False;
prefix = "Completed";
suffix = "Node Crawls";
completed_node_count = 0;
while not self.stopreuest.isSet():
if not self.progress_queue.empty():
completed_node_count = self.progress_queue.get();
while not self.progress_queue.empty():
completed_node_count = self.progress_queue.get();
if not message_printed:
print "%s %5d %s"%(prefix,
completed_node_count,
suffix),
# This is because the crawl count
# is given 10 positions and we have a space
print "\b"*(8 + len(suffix)),
sys.stdout.flush();
message_printed = True;
else:
print "%5d"%(completed_node_count),
print "\b"*7,
sys.stdout.flush();
time.sleep(self.refreash_rate);
print "%6d %s\nGraph Collected at: %s"%(completed_node_count,
suffix,
self.output_file_name),
sys.stdout.flush();
def crawl(self):
my_filter = GraphURLFilter(
[entry['URL'].strip() for entry in self.input_hash]);
node_count = 0;
self.progress_queue.put(node_count);
self.progress_bar_thread.start();
for entry in self.input_hash:
TemporalPageCrawler.TemporalPageCrawlerThreaded(
entry['URL'].strip(),
my_filter,
self.output_file_name).get_temporal_crawl();
node_count += 1;
self.progress_queue.put(node_count);
# Job Completed Close the progress bar gracefully
self.stopreuest.set();
# waiting for progress bar to wake up and print the close message
self.progress_bar_thread.join(self.refreash_rate+10);
if self.progress_bar_thread.isAlive():
# Trouble in paradise
self.progress_bar_thread.stop();
class GraphCrawlerSimple:
"""
Grap
"""
def __init__(self, input_file_name, file_delimiter = "\t", start_index = 0, end_index = 2):
self.file_delimiter = file_delimiter;
self.input_hash = self.process_csv_file(input_file_name,
file_delimiter);
self.output_file_name = self.__make_output_file_name(
input_file_name);
self.start_index = start_index;
self.end_index = end_index;
def __make_output_file_name(self, input_file_name):
fileName, fileExtension = os.path.splitext(input_file_name);
return fileName + self.start_index + "_" + self.end_index + ".out";
def process_csv_file(self, in_file_name, delimiter):
result_hash = [];
with open(in_file_name, 'rb') as csv_file:
inp_stream = csv.DictReader(csv_file,
delimiter = delimiter);
for entry in inp_stream:
result_hash.append(entry);
return result_hash;
def crawl(self):
my_filter = GraphURLFilter(
[entry['URL'].strip() for entry in self.input_hash]);
node_count = 0;
for entry in self.input_hash:
if self.start_index < node_count:
node_count += 1;
continue;
if self.end_index > node_count:
break;
TemporalPageCrawler.TemporalPageCrawlerThreaded(
entry['URL'].strip(),
my_filter,
self.output_file_name).get_temporal_crawl();
node_count += 1;
def process(self):
self.crawl()
class GraphCrawlerMultiProcess:
def __init__(self, input_file_name, file_delimiter='\t'):
self.input_file_name = input_file_name;
self.file_delimiter = file_delimiter;
def node_count(self, in_file_name, delimiter):
result_hash = [];
with open(in_file_name, 'rb') as csv_file:
inp_stream = csv.DictReader(csv_file,
delimiter = delimiter);
for entry in inp_stream:
result_hash.append(entry);
return len(result_hash);
def process(self):
node_count_max = self.node_count(self.input_file_name, self.file_delimiter);
node_count = 0;
jump = 4;
pm = process_manager.ProcessManager();
pm.start_workers();
while node_count < node_count_max:
gc = GraphCrawlerSimple(self.input_file_name,
self.file_delimiter,
node_count,
node_count + jump - 1)
pm.add_jobs(gc)
node_count += jump
if wait:
while not pm.are_all_jobs_completed():
time.sleep(500);
pm.close();
return;
if __name__ == "__main__":
# Crawling over test set
GraphCrawlerMultiProcess(sys.argv[1]).crawl();