gendummydata.py 8.7 KB
Newer Older
1
#!/usr/bin/python2
eric's avatar
eric committed
2
3
4
"""
usage: gendummydata.py outputfilename.sql
"""
eric's avatar
eric committed
5
6
7
8
9
10
11
12
13
14
15
16
17
18
#
# This script seeds the AUR database with dummy data for
# use during development/testing.  It uses random entries
# from /usr/share/dict/words to create user accounts and
# package names.  It generates the SQL statements to
# insert these users/packages into the AUR database.
#

DBUG      = 1
SEED_FILE = "/usr/share/dict/words"
DB_HOST   = "localhost"
DB_NAME   = "AUR"
DB_USER   = "aur"
DB_PASS   = "aur"
eric's avatar
eric committed
19
USER_ID   = 5          # Users.ID of first bogus user
eric's avatar
eric committed
20
PKG_ID    = 1          # Packages.ID of first package
Loui Chang's avatar
Loui Chang committed
21
MAX_USERS = 300        # how many users to 'register'
eric's avatar
eric committed
22
23
MAX_DEVS  = .1         # what percentage of MAX_USERS are Developers
MAX_TUS   = .2         # what percentage of MAX_USERS are Trusted Users
Loui Chang's avatar
Loui Chang committed
24
MAX_PKGS  = 900       # how many packages to load
eric's avatar
eric committed
25
PKG_FILES = (8, 30)    # min/max number of files in a package
26
27
PKG_DEPS  = (1, 5)     # min/max depends a package has
PKG_SRC   = (1, 3)     # min/max sources a package has
eric's avatar
eric committed
28
PKG_CMNTS = (1, 5)     # min/max number of comments a package has
eric's avatar
eric committed
29
VOTING    = (0, .30)   # percentage range for package voting
eric's avatar
eric committed
30
RANDOM_PATHS = (       # random path locations for package files
eric's avatar
eric committed
31
32
33
34
	"/usr/bin", "/usr/lib", "/etc", "/etc/rc.d", "/usr/share", "/lib",
	"/var/spool", "/var/log", "/usr/sbin", "/opt", "/usr/X11R6/bin",
	"/usr/X11R6/lib", "/usr/libexec", "/usr/man/man1", "/usr/man/man3",
	"/usr/man/man5", "/usr/X11R6/man/man1", "/etc/profile.d"
eric's avatar
eric committed
35
36
37
38
)
RANDOM_TLDS = ("edu", "com", "org", "net", "tw", "ru", "pl", "de", "es")
RANDOM_URL = ("http://www.", "ftp://ftp.", "http://", "ftp://")
RANDOM_LOCS = ("pub", "release", "files", "downloads", "src")
Loui Chang's avatar
Loui Chang committed
39
FORTUNE_CMD = "/usr/bin/fortune -l"
40

eric's avatar
eric committed
41
42
43
44
45
46

import random
import time
import os
import sys
import cStringIO
eric's avatar
eric committed
47
48
import commands

eric's avatar
eric committed
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

if len(sys.argv) != 2:
	sys.stderr.write("Missing output filename argument");
	raise SystemExit

# Just let python throw the errors if any happen
#
out = open(sys.argv[1], "w")

# make sure the seed file exists
#
if not os.path.exists(SEED_FILE):
	sys.stderr.write("Please install the 'words' Arch package\n");

# Make sure database access will be available
#
try:
	import MySQLdb
except:
	sys.stderr.write("Please install the 'mysql-python' Arch package\n");
	raise SystemExit

# try to connect to database
#
try:
	db = MySQLdb.connect(host = DB_HOST, user = DB_USER,
			db = DB_NAME, passwd = DB_PASS)
	dbc = db.cursor()
except:
	sys.stderr.write("Could not connect to database\n");
	raise SystemExit

eric's avatar
eric committed
81
82
esc = db.escape_string

eric's avatar
eric committed
83
84
85
86
87
88
89
90
91
92
93

# track what users/package names have been used
#
seen_users = {}
seen_pkgs = {}
categories = {}
category_keys = []
user_keys = []

# some functions to generate random data
#
94
def genVersion():
eric's avatar
eric committed
95
96
97
	ver = []
	ver.append("%d" % random.randrange(0,10))
	ver.append("%d" % random.randrange(0,20))
eric's avatar
eric committed
98
	if random.randrange(0,2) == 0:
eric's avatar
eric committed
99
		ver.append("%d" % random.randrange(0,100))
100
	return ".".join(ver) + "-u%d" % random.randrange(1,11)
eric's avatar
eric committed
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
def genCategory():
	return categories[category_keys[random.randrange(0,len(category_keys))]]
def genUID():
	return seen_users[user_keys[random.randrange(0,len(user_keys))]]


# load the words, and make sure there are enough words for users/pkgs
#
if DBUG: print "Grabbing words from seed file..."
fp = open(SEED_FILE, "r")
contents = fp.readlines()
fp.close()
if MAX_USERS > len(contents):
	MAX_USERS = len(contents)
if MAX_PKGS > len(contents):
	MAX_PKGS = len(contents)
if len(contents) - MAX_USERS > MAX_PKGS:
	need_dupes = 0
else:
	need_dupes = 1

# select random usernames
#
if DBUG: print "Generating random user names..."
user_id = USER_ID
while len(seen_users) < MAX_USERS:
	user = random.randrange(0, len(contents))
eric's avatar
eric committed
128
129
130
	word = contents[user].replace("'", "").replace(".","").replace(" ", "_")
	word = word.strip().lower()
	if not seen_users.has_key(word):
eric's avatar
eric committed
131
132
133
134
135
136
137
138
139
140
		seen_users[word] = user_id
		user_id += 1
user_keys = seen_users.keys()

# select random package names
#
if DBUG: print "Generating random package names..."
num_pkgs = PKG_ID
while len(seen_pkgs) < MAX_PKGS:
	pkg = random.randrange(0, len(contents))
eric's avatar
eric committed
141
142
	word = contents[pkg].replace("'", "").replace(".","").replace(" ", "_")
	word = word.strip().lower()
eric's avatar
eric committed
143
144
145
146
147
148
149
150
151
	if not need_dupes:
		if not seen_pkgs.has_key(word) and not seen_users.has_key(word):
			seen_pkgs[word] = num_pkgs
			num_pkgs += 1
	else:
		if not seen_pkgs.has_key(word):
			seen_pkgs[word] = num_pkgs
			num_pkgs += 1

eric's avatar
eric committed
152
153
154
155
# free up contents memory
#
contents = None

eric's avatar
eric committed
156
157
# Load package categories from database
#
158
if DBUG: print "Loading package categories..."
eric's avatar
eric committed
159
160
161
162
163
164
165
166
167
168
169
170
171
q = "SELECT * FROM PackageCategories"
dbc.execute(q)
row = dbc.fetchone()
while row:
	categories[row[1]] = row[0]
	row = dbc.fetchone()
category_keys = categories.keys()

# done with the database
#
dbc.close()
db.close()

172
173
174
175
176
177
178
# developer/tu IDs
#
developers = []
trustedusers = []
has_devs = 0
has_tus = 0

eric's avatar
eric committed
179
180
181
182
183
# Begin by creating the User statements
#
if DBUG: print "Creating SQL statements for users.",
count = 0
for u in user_keys:
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
	account_type = 1  # default to normal user
	if not has_devs or not has_tus:
		account_type = random.randrange(1, 4)
		if account_type == 3 and not has_devs:
			# this will be a dev account
			#
			developers.append(seen_users[u])
			if len(developers) >= MAX_DEVS * MAX_USERS:
				has_devs = 1
		elif account_type == 2 and not has_tus:
			# this will be a trusted user account
			#
			trustedusers.append(seen_users[u])
			if len(trustedusers) >= MAX_TUS * MAX_USERS:
				has_tus = 1
		else:
			# a normal user account
			#
			pass
	
204
	s = "INSERT INTO Users (ID, AccountTypeID, Username, Email, Passwd) VALUES (%d, %d, '%s', '%s@example.com', MD5('%s'));\n" % (seen_users[u], account_type, u, u, u)
eric's avatar
eric committed
205
206
207
208
209
	out.write(s)
	if count % 10 == 0:
		if DBUG: print ".",
	count += 1
if DBUG: print "."
210
211
212
213
214
if DBUG:
	print "Number of developers:", len(developers)
	print "Number of trusted users:", len(trustedusers)
	print "Number of users:", (MAX_USERS-len(developers)-len(trustedusers))
	print "Number of packages:", MAX_PKGS
eric's avatar
eric committed
215
216
217
218
219
220

# Create the package statements
#
if DBUG: print "Creating SQL statements for packages.",
count = 0
for p in seen_pkgs.keys():
eric's avatar
eric committed
221
	NOW = int(time.time())
222
223
224
225
226
227
228
	if count % 2 == 0:
		muid = developers[random.randrange(0,len(developers))]
	else:
		muid = trustedusers[random.randrange(0,len(trustedusers))]
	if count % 20 == 0: # every so often, there are orphans...
		muid = 0

eric's avatar
eric committed
229
230
	uuid = genUID() # the submitter/user

231
232
	s = "INSERT INTO Packages (ID, Name, Version, CategoryID, SubmittedTS, SubmitterUID, MaintainerUID) VALUES (%d, '%s', '%s', %d, %d, %d, %d);\n" % (seen_pkgs[p], p, genVersion(),
			genCategory(), NOW, uuid, muid)
eric's avatar
eric committed
233
234
235
236
237
	out.write(s)
	if count % 100 == 0:
		if DBUG: print ".",
	count += 1

eric's avatar
eric committed
238
239
240
241
	# create random comments for this package
	#
	num_comments = random.randrange(PKG_CMNTS[0], PKG_CMNTS[1])
	for i in range(0, num_comments):
eric's avatar
eric committed
242
		fortune = esc(commands.getoutput(FORTUNE_CMD).replace("'",""))
eric's avatar
eric committed
243
		now = NOW + random.randrange(400, 86400*3)
eric's avatar
eric committed
244
		s = "INSERT INTO PackageComments (PackageID, UsersID, Comments, CommentTS) VALUES (%d, %d, '%s', %d);\n" % (seen_pkgs[p], genUID(), fortune, now)
eric's avatar
eric committed
245
246
		out.write(s)

eric's avatar
eric committed
247
248
249
250
if DBUG: print "."

# Cast votes
#
251
track_votes = {}
eric's avatar
eric committed
252
253
254
if DBUG: print "Casting votes for packages.",
count = 0
for u in user_keys:
255
256
	num_votes = random.randrange(int(len(seen_pkgs)*VOTING[0]),
			int(len(seen_pkgs)*VOTING[1]))
eric's avatar
eric committed
257
258
259
260
	pkgvote = {}
	for v in range(num_votes):
		pkg = random.randrange(0, len(seen_pkgs))
		if not pkgvote.has_key(pkg):
eric's avatar
eric committed
261
			s = "INSERT INTO PackageVotes (UsersID, PackageID) VALUES (%d, %d);\n" % (seen_users[u], pkg)
eric's avatar
eric committed
262
			pkgvote[pkg] = 1
263
264
265
			if not track_votes.has_key(pkg):
				track_votes[pkg] = 0
			track_votes[pkg] += 1
eric's avatar
eric committed
266
267
268
269
270
			out.write(s)
			if count % 100 == 0:
				if DBUG: print ".",
			count += 1

271
272
273
274
275
276
# Update statements for package votes
#
for p in track_votes.keys():
	s = "UPDATE Packages SET NumVotes = %d WHERE ID = %d;\n" % (track_votes[p], p)
	out.write(s)

277
278
279
280
281
282
283
284
285
286
287
# Create package dependencies and sources
#
if DBUG: print "."; print "Creating statements for package depends/sources.",
count = 0
for p in seen_pkgs.keys():
	num_deps = random.randrange(PKG_DEPS[0], PKG_DEPS[1])
	this_deps = {}
	i = 0
	while i != num_deps:
		dep = random.randrange(0, len(seen_pkgs))
		if not this_deps.has_key(dep):
Loui Chang's avatar
Loui Chang committed
288
			s = "INSERT INTO PackageDepends VALUES (%d, %d, NULL);\n" % (seen_pkgs[p], dep)
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
			out.write(s)
			i += 1

	num_sources = random.randrange(PKG_SRC[0], PKG_SRC[1])
	for i in range(num_sources):
		src_file = user_keys[random.randrange(0, len(user_keys))]
		src = "%s%s.%s/%s/%s-%s.tar.gz" % (
				RANDOM_URL[random.randrange(0,len(RANDOM_URL))],
				p, RANDOM_TLDS[random.randrange(0,len(RANDOM_TLDS))],
				RANDOM_LOCS[random.randrange(0,len(RANDOM_LOCS))],
				src_file, genVersion())
		s = "INSERT INTO PackageSources VALUES (%d, '%s');\n" % (
				seen_pkgs[p], src)
		out.write(s)

	if count % 100 == 0:
		if DBUG: print ".",
	count += 1


eric's avatar
eric committed
309
310
311
312
313
314
315
316
# close output file
#
out.write("\n")
out.close()

if DBUG: print "."
if DBUG: print "Done."