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
252
253
254
255
256
257
258
259
|
#!/usr/bin/python
#
# Author: Brian Murray <brian@canonical.com>
# Copyright (C) 2011 Canonical, Ltd.
# License: GPLv3
#
# Using a local copy of the UDD (Ultimate Debian Database) check to find bug
# reports with similar titles about Ubuntu and or Debian
from launchpadlib.errors import HTTPError
from operator import itemgetter
import datetime
import lpl_common
import Levenshtein
import optparse
import psycopg2
def connect_to_db():
host = 'localhost'
dbname = 'udd'
user = 'udd'
connect = psycopg2.connect("host=%s dbname=%s user=%s password=password" %
(host, dbname, user))
return connect.cursor()
def prepare_link(row):
if row[1][1] == 'launchpad':
return "LP: #%s" % row[0]
elif row[1][1] == 'debian':
return "http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=%s" % row[0]
def check_lp_skippable(number):
''' Skip bugs that duplicates, private or have debian bug watches.'''
bug = lp.bugs[number]
try:
bug.duplicate_of
except HTTPError:
# skip private bugs
return True
except ValueError:
return True
if bug.duplicate_of:
return True
if len([bug_watch for bug_watch in bug.bug_watches]) > 0:
if 'bugs.debian.org' in str([bug_watch.url for bug_watch in bug.bug_watches]):
#if 'debbugs' in [bug_watch.bug_tracker.name for bug_watch in bug.bug_watches]:
# if the bug has a debian bug watch skip it
return True
else:
return False
lp = lpl_common.connect(use_edge=False, version='devel')
cur = connect_to_db()
parser = optparse.OptionParser(usage="usage %prog --package PACKAGE \
--fixed")
parser.add_option('--package', help='source package name')
parser.add_option('--fixed', help='Search archived debian bugs',
default=False, action="store_true")
parser.add_option('--team',
help='Display bugs about packages TEAM is subscribed to')
parser.add_option('--within',
help='Compare bugs created since X days ago')
parser.add_option('--different',
help='Only show bugs from different bug trackers',
default=False, action="store_true")
(opt, args) = parser.parse_args()
package = opt.package
ubuntu = lp.projects['ubuntu']
if opt.team:
team = lp.people[opt.team]
pkgs = team.getBugSubscriberPackages()
else:
pkgs = [ubuntu.getSourcePackage(name=package)]
for package in pkgs:
pkg_name = package.name
# bug_tracker, bug_number, title
# 2011-09-30: need to filter on some date thing to make it ones changed
# since a certain date
cur.execute('''SELECT 'launchpad', ubuntu_bugs.bug, ubuntu_bugs.title, ubuntu_bugs.date_reported
FROM ubuntu_bugs INNER JOIN ubuntu_bugs_tasks
ON ubuntu_bugs.bug = ubuntu_bugs_tasks.bug
WHERE ubuntu_bugs.duplicate_of is NULL
AND ubuntu_bugs_tasks.package = '%s' AND
ubuntu_bugs.bug NOT IN
(SELECT ubuntu_bugs_tags.bug FROM ubuntu_bugs_tags
WHERE ubuntu_bugs_tags.tag = 'apport-package')'''
% pkg_name)
results = cur.fetchall()
if not opt.fixed:
# the merged bit was taken from an example query at the UDD page and
# is suspect as I don't understand it exactly
cur.execute('''SELECT 'debian', bugs.id, bugs.title, bugs.arrival
FROM bugs WHERE
source = '%s'
AND (NOT (bugs.id IN
(SELECT DISTINCT bugs_merged_with.id FROM bugs_merged_with
WHERE (bugs_merged_with.id > bugs_merged_with.merged_with))
))''' % pkg_name)
elif opt.fixed:
cur.execute('''SELECT 'debian', archived_bugs.id, archived_bugs.title
FROM archived_bugs WHERE source = '%s' ''' % pkg_name)
deb_results = cur.fetchall()
# do some pre-cleaning of debian titles
for deb_result in deb_results:
if 'intl:' in deb_result[2].lower() or \
'translation' in deb_result[2].lower():
continue
else:
results.append(deb_result)
bugs = {}
for result in results:
tracker = result[0]
# if you don't remove the space there would be a double space
releases = ['breezy ', 'dapper ', 'edgy ', 'feisty ', 'gutsy ',
'hardy ', 'intrepid ', 'jaunty ', 'karmic ', 'lucid ',
'maverick ', 'natty ', 'oneiric ']
bug_number = result[1]
# lower case all of the titles
try:
title = result[2].lower()
except AttributeError:
continue
# perhaps using regex would be better
title = title.strip('''"''')
title = title.strip("'")
# frequently seen with release
title = title.strip('[')
# frequently seen with release
title = title.replace(']', '')
# make contracted words uniform
title = title.replace('''can't''', 'cannot')
title = title.replace('''doesn't''', 'does not')
title = title.replace('''isn't''', 'is not')
# release in title is not useful
for release in releases:
title = title.replace(release, '')
# package in start of title is redundant
try:
if title.startswith('%s ' % package):
title = title.replace('%s ' % package, '')
except UnicodeDecodeError:
continue
try:
if title.startswith('%s: ' % package):
title = title.replace('%s: ' % package, '')
except UnicodeDecodeError:
continue
title = title.strip()
# lp date created is not a datetime so make it one
if tracker == 'launchpad':
created = datetime.datetime.strptime(result[3].replace(' +0000', ''), '%a, %d %b %Y %H:%M:%S')
else:
created = result[3]
bugs[bug_number] = (title, tracker, created)
sorted_titles = sorted(bugs.items(), key=itemgetter(1))
if opt.within:
today = datetime.datetime.now()
target_date = today - datetime.timedelta(int(opt.within))
else:
target_date = None
start = 0
while start < len(sorted_titles)-1:
after = start + 1
if target_date:
if sorted_titles[start][1][2] < target_date or sorted_titles[start][1][2] < target_date:
start += 1
continue
if opt.different:
if sorted_titles[start][1][1] == sorted_titles[after][1][1]:
start += 1
continue
this_title = sorted_titles[start][1][0]
next_title = sorted_titles[after][1][0]
if this_title == '' or this_title == '(no subject)':
start += 1
continue
if sorted_titles[start][1][1] == 'launchpad' and \
check_lp_skippable(sorted_titles[start][0]) is True:
start += 1
continue
elif sorted_titles[after][1][1] == 'launchpad' and \
check_lp_skippable(sorted_titles[after][0]) is True:
# actually shouldn't this be plus 2? because we are on the next title already
start += 1
continue
try:
ratio = Levenshtein.ratio(this_title, next_title)
except TypeError:
start += 1
continue
if ratio != 1.00 and 'sigsegv' in this_title:
start += 1
continue
if ratio != 1.00 and 'assert failure' in this_title:
start += 1
continue
if ratio > 0.75 and not 'sigsegv' in this_title:
slink = prepare_link(sorted_titles[start])
alink = prepare_link(sorted_titles[after])
print "%s" % ratio
print "\t%s: %s" % (slink, sorted_titles[start][1][0])
print "\t%s: %s" % (alink, sorted_titles[after][1][0])
match = True
increment = 1
if after + increment >= len(sorted_titles):
start += 1
continue
while match is True:
next_next_title = sorted_titles[after+increment][1][0]
if next_next_title == '' or \
next_next_title == '(no subject)':
match = False
start += 1
continue
nratio = Levenshtein.ratio(next_title, next_next_title)
if nratio == ratio:
nlink = prepare_link(sorted_titles[after+increment])
print "\t%s: %s" % \
(nlink, sorted_titles[after+increment][1][0])
increment += 1
else:
# need to increase start here to account for the fact that we've gone further in the results set
start += increment - 1
match = False
start += 1
elif ratio == 1.00 and 'sigsegv' in this_title:
slink = prepare_link(sorted_titles[start])
alink = prepare_link(sorted_titles[after])
print "%s" % ratio
print "\t%s: %s" % (slink, sorted_titles[start][1][0])
print "\t%s: %s" % (alink, sorted_titles[after][1][0])
start += 1
elif ratio == 1.00 and 'assert failure' in this_title:
slink = prepare_link(sorted_titles[start])
alink = prepare_link(sorted_titles[after])
print "%s" % ratio
print "\t%s: %s" % (slink, sorted_titles[start][1][0])
print "\t%s: %s" % (alink, sorted_titles[after][1][0])
start += 1
start += 1
|