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
|
#! /usr/bin/python
#
# Tests inserting and querying with huge amounts of events
#
# Update python path to use local zeitgeist module
import sys
import signal
import os
sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
from subprocess import Popen, PIPE
from tempfile import mkstemp
from zeitgeist.datamodel import *
from zeitgeist.client import *
REPORT = \
"""EVENT METADATA:
Number of events: %s
Number of event interpretations: %s
Number of event manifestations: %s
Number of event actors: %s
Event payload frequency: %s
Event subject frequency: %s
SUBJECT METADATA:
Number of subjects: %s
Number if subject interpretations: %s
Number of subject manifestations: %s
Number if subject mimetypes: %s
Number of subject origins: %s
Number of subject storage mediums: %s
Subject text frequency: %s"""
class EventGenerator:
"""
Generate a collection of random events. The entire event set is
pre-compiled in order to not influence benchmarks etc. with
event generation time.
"""
def __init__ (self,
num_events=100000,
num_event_interpretations=10,
num_event_manifestations=5,
num_event_actors=100,
num_subjects=10000,
num_subject_interpretations=100,
num_subject_manifestations=50,
num_subject_mimetypes=50,
num_subject_origins=70000,
num_subject_storages=8,
subject_freq=1.0,
subject_text_freq=0.0,
payload_freq=0.0):
self.subject_freq = subject_freq
self.subject_text_freq = subject_text_freq
self.payload_freq = payload_freq
# Event metadata
self.event_interpretations = ["interpretation%s" % i for i in range(num_event_interpretations)]
self.event_manifestations = ["manifestation%s" % i for i in range(num_event_manifestations)]
self.event_actors = ["actor%s" % i for i in range(num_event_actors)]
# Subject data
self.subject_uris = ["subject%s" % i for i in range(num_subjects)]
self.subject_interpretations = ["subject_interpretation%s" % i for i in range(num_subject_interpretations)]
self.subject_manifestations = ["subject_manifestation%s" % i for i in range(num_subject_manifestations)]
self.subject_mimetypes = ["subject_mimetype%s" % i for i in range(num_subject_mimetypes)]
self.subject_origins = ["subject_origin%s" % i for i in range(num_subject_origins)]
self.subject_storages = ["subject_storage%s" % i for i in range(num_subject_storages)]
# Compile all the events ahead of time in order not to
# influence query/insertion time.
# We give each event a bogus timestamp in order to avoid
# duplicate event exceptions
self.events = []
for i in range(num_events):
ev = Event()
ev.timestamp = i
ev.interpretation = self.event_interpretations[i % len(self.event_interpretations)]
ev.manifestation = self.event_manifestations[i % len(self.event_manifestations)]
ev.actor = self.event_actors[i % len(self.event_actors)]
#if payload_freq > 0 and (int((1/payload_freq)*num_events) % i == 0):
# event.payload = "payload%s" % i
for j in range(self._calc_num_subjects()):
subj = Subject()
subj.uri = self.subject_uris[(i+j) % num_subjects]
subj.interpretation = self.subject_interpretations[(i+j) % len(self.subject_interpretations)]
subj.manifestation = self.subject_manifestations[(i+j) % len(self.subject_manifestations)]
subj.mimetype = self.subject_mimetypes[(i+j) % len(self.subject_mimetypes)]
subj.origin = self.subject_origins[(i+j) % len(self.subject_origins)]
subj.storage = self.subject_storages[(i+j) % len(self.subject_storages)]
#if subject_text_freq > 0 and (int((1/subject_text_freq)*num_subjects) % (i+j) == 0):
# event.payload = "payload%s" % i
ev.subjects.append(subj)
self.events.append(ev)
def __iter__ (self):
return self.events.__iter__()
def __len__ (self):
return self.events.__len__()
def __getitem__ (self, key):
return self.events[key]
def _calc_num_subjects(self):
# FIXME: DO this right
return 1
def report(self):
return REPORT % (len(self), len(self.event_interpretations), len(self.event_manifestations), len(self.event_actors), self.payload_freq, self.subject_freq, len(self.subject_uris), len(self.subject_interpretations), len(self.subject_manifestations), len(self.subject_mimetypes), len(self.subject_origins), len(self.subject_storages), self.subject_text_freq)
def spawn_daemon(database_file, logger=PIPE):
os.environ.update({"ZEITGEIST_DATABASE_PATH": database_file})
daemon = Popen(
["./zeitgeist-daemon.py", "--no-datahub"], stderr=logger, stdout=logger
)
time.sleep(3)
err = daemon.poll()
if err:
raise RuntimeError("Could not start daemon, got err=%i" % err)
return daemon
def kill_daemon(daemon):
try:
os.kill(daemon.pid, signal.SIGKILL)
except (OSError, AttributeError):
pass
else:
daemon.wait()
if __name__ == "__main__":
import sys, time
if len(sys.argv) == 2:
logger = open(sys.argv[-1], "w")
else:
logger=PIPE
events = EventGenerator(num_events=20000, num_subjects=500)
print events.report() + "\n"
fd, database_file = mkstemp(prefix="zeitgeist.", suffix=".sqlite")
print "DATABASE:", database_file
os.close(fd)
daemon = spawn_daemon(database_file, logger)
try:
log = ZeitgeistDBusInterface()
start = time.time()
# Insert events in batches of 10
for i in range(len(events) / 10):
batch = [events[i*10 + j] for j in range(10)]
log.InsertEvents(batch)
print "Insertion time: %ss" % (time.time() - start)
# test for lp #583065
templates = []
for uri in events.subject_uris[:30]:
templates.append(Event.new_for_values(subjects=[Subject.new_for_values(uri=uri)]))
start = time.time()
ev = log.FindEvents(
TimeRange.until_now(),
templates,
StorageState.Any,
0, #all possible results
ResultType.MostPopularSubjects
)
print "Query time: %ss" % (time.time() - start)
finally:
kill_daemon(daemon)
# comment the next line if you would like to kkep the database
os.unlink(database_file)
|