summaryrefslogtreecommitdiffstats
path: root/Tools/Scripts/webkitpy/common/net/bugzilla/bugzilla.py
blob: 8daf92e2e808c696efef6bdce78f79244e276f65 (plain)
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
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
# Copyright (c) 2009 Google Inc. All rights reserved.
# Copyright (c) 2009 Apple Inc. All rights reserved.
# Copyright (c) 2010 Research In Motion Limited. All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met:
#
#     * Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
#     * Redistributions in binary form must reproduce the above
# copyright notice, this list of conditions and the following disclaimer
# in the documentation and/or other materials provided with the
# distribution.
#     * Neither the name of Google Inc. nor the names of its
# contributors may be used to endorse or promote products derived from
# this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# WebKit's Python module for interacting with Bugzilla

import mimetypes
import os.path
import re
import StringIO
import urllib

from datetime import datetime # used in timestamp()

from .attachment import Attachment
from .bug import Bug

from webkitpy.common.system.deprecated_logging import log
from webkitpy.common.config import committers
from webkitpy.common.net.credentials import Credentials
from webkitpy.common.system.user import User
from webkitpy.thirdparty.autoinstalled.mechanize import Browser
from webkitpy.thirdparty.BeautifulSoup import BeautifulSoup, SoupStrainer


# FIXME: parse_bug_id should not be a free function.
def parse_bug_id(message):
    if not message:
        return None
    match = re.search(Bugzilla.bug_url_short, message)
    if match:
        return int(match.group('bug_id'))
    match = re.search(Bugzilla.bug_url_long, message)
    if match:
        return int(match.group('bug_id'))
    return None


# FIXME: parse_bug_id_from_changelog should not be a free function.
# Parse the bug ID out of a Changelog message based on the format that is
# used by prepare-ChangeLog
def parse_bug_id_from_changelog(message):
    if not message:
        return None
    match = re.search("^\s*" + Bugzilla.bug_url_short + "$", message, re.MULTILINE)
    if match:
        return int(match.group('bug_id'))
    match = re.search("^\s*" + Bugzilla.bug_url_long + "$", message, re.MULTILINE)
    if match:
        return int(match.group('bug_id'))
    return None

def timestamp():
    return datetime.now().strftime("%Y%m%d%H%M%S")


# A container for all of the logic for making and parsing bugzilla queries.
class BugzillaQueries(object):

    def __init__(self, bugzilla):
        self._bugzilla = bugzilla

    def _is_xml_bugs_form(self, form):
        # ClientForm.HTMLForm.find_control throws if the control is not found,
        # so we do a manual search instead:
        return "xml" in [control.id for control in form.controls]

    # This is kinda a hack.  There is probably a better way to get this information from bugzilla.
    def _parse_result_count(self, results_page):
        result_count_text = BeautifulSoup(results_page).find(attrs={'class': 'bz_result_count'}).string
        result_count_parts = result_count_text.strip().split(" ")
        if result_count_parts[0] == "Zarro":
            return 0
        if result_count_parts[0] == "One":
            return 1
        return int(result_count_parts[0])

    # Note: _load_query, _fetch_bug and _fetch_bugs_from_advanced_query
    # are the only methods which access self._bugzilla.

    def _load_query(self, query):
        self._bugzilla.authenticate()
        full_url = "%s%s" % (self._bugzilla.bug_server_url, query)
        return self._bugzilla.browser.open(full_url)

    def _fetch_bugs_from_advanced_query(self, query):
        results_page = self._load_query(query)
        if not self._parse_result_count(results_page):
            return []
        # Bugzilla results pages have an "XML" submit button at the bottom
        # which can be used to get an XML page containing all of the <bug> elements.
        # This is slighty lame that this assumes that _load_query used
        # self._bugzilla.browser and that it's in an acceptable state.
        self._bugzilla.browser.select_form(predicate=self._is_xml_bugs_form)
        bugs_xml = self._bugzilla.browser.submit()
        return self._bugzilla._parse_bugs_from_xml(bugs_xml)

    def _fetch_bug(self, bug_id):
        return self._bugzilla.fetch_bug(bug_id)

    def _fetch_bug_ids_advanced_query(self, query):
        soup = BeautifulSoup(self._load_query(query))
        # The contents of the <a> inside the cells in the first column happen
        # to be the bug id.
        return [int(bug_link_cell.find("a").string)
                for bug_link_cell in soup('td', "first-child")]

    def _parse_attachment_ids_request_query(self, page):
        digits = re.compile("\d+")
        attachment_href = re.compile("attachment.cgi\?id=\d+&action=review")
        attachment_links = SoupStrainer("a", href=attachment_href)
        return [int(digits.search(tag["href"]).group(0))
                for tag in BeautifulSoup(page, parseOnlyThese=attachment_links)]

    def _fetch_attachment_ids_request_query(self, query):
        return self._parse_attachment_ids_request_query(self._load_query(query))

    def _parse_quips(self, page):
        soup = BeautifulSoup(page, convertEntities=BeautifulSoup.HTML_ENTITIES)
        quips = soup.find(text=re.compile(r"Existing quips:")).findNext("ul").findAll("li")
        return [unicode(quip_entry.string) for quip_entry in quips]

    def fetch_quips(self):
        return self._parse_quips(self._load_query("/quips.cgi?action=show"))

    # List of all r+'d bugs.
    def fetch_bug_ids_from_pending_commit_list(self):
        needs_commit_query_url = "buglist.cgi?query_format=advanced&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&field0-0-0=flagtypes.name&type0-0-0=equals&value0-0-0=review%2B"
        return self._fetch_bug_ids_advanced_query(needs_commit_query_url)

    def fetch_bugs_matching_quicksearch(self, search_string):
        # We may want to use a more explicit query than "quicksearch".
        # If quicksearch changes we should probably change to use
        # a normal buglist.cgi?query_format=advanced query.
        quicksearch_url = "buglist.cgi?quicksearch=%s" % urllib.quote(search_string)
        return self._fetch_bugs_from_advanced_query(quicksearch_url)

    # Currently this returns all bugs across all components.
    # In the future we may wish to extend this API to construct more restricted searches.
    def fetch_bugs_matching_search(self, search_string, author_email=None):
        query = "buglist.cgi?query_format=advanced"
        if search_string:
            query += "&short_desc_type=allwordssubstr&short_desc=%s" % urllib.quote(search_string)
        if author_email:
            query += "&emailreporter1=1&emailtype1=substring&email1=%s" % urllib.quote(search_string)
        return self._fetch_bugs_from_advanced_query(query)

    def fetch_patches_from_pending_commit_list(self):
        return sum([self._fetch_bug(bug_id).reviewed_patches()
            for bug_id in self.fetch_bug_ids_from_pending_commit_list()], [])

    def fetch_bug_ids_from_commit_queue(self):
        commit_queue_url = "buglist.cgi?query_format=advanced&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&field0-0-0=flagtypes.name&type0-0-0=equals&value0-0-0=commit-queue%2B&order=Last+Changed"
        return self._fetch_bug_ids_advanced_query(commit_queue_url)

    def fetch_patches_from_commit_queue(self):
        # This function will only return patches which have valid committers
        # set.  It won't reject patches with invalid committers/reviewers.
        return sum([self._fetch_bug(bug_id).commit_queued_patches()
                    for bug_id in self.fetch_bug_ids_from_commit_queue()], [])

    def fetch_bug_ids_from_review_queue(self):
        review_queue_url = "buglist.cgi?query_format=advanced&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&field0-0-0=flagtypes.name&type0-0-0=equals&value0-0-0=review?"
        return self._fetch_bug_ids_advanced_query(review_queue_url)

    # This method will make several requests to bugzilla.
    def fetch_patches_from_review_queue(self, limit=None):
        # [:None] returns the whole array.
        return sum([self._fetch_bug(bug_id).unreviewed_patches()
            for bug_id in self.fetch_bug_ids_from_review_queue()[:limit]], [])

    # NOTE: This is the only client of _fetch_attachment_ids_request_query
    # This method only makes one request to bugzilla.
    def fetch_attachment_ids_from_review_queue(self):
        review_queue_url = "request.cgi?action=queue&type=review&group=type"
        return self._fetch_attachment_ids_request_query(review_queue_url)


class Bugzilla(object):

    def __init__(self, dryrun=False, committers=committers.CommitterList()):
        self.dryrun = dryrun
        self.authenticated = False
        self.queries = BugzillaQueries(self)
        self.committers = committers
        self.cached_quips = []

        # FIXME: We should use some sort of Browser mock object when in dryrun
        # mode (to prevent any mistakes).
        self.browser = Browser()
        # Ignore bugs.webkit.org/robots.txt until we fix it to allow this
        # script.
        self.browser.set_handle_robots(False)

    # FIXME: Much of this should go into some sort of config module:
    bug_server_host = "bugs.webkit.org"
    bug_server_regex = "https?://%s/" % re.sub('\.', '\\.', bug_server_host)
    bug_server_url = "https://%s/" % bug_server_host
    bug_url_long = bug_server_regex + r"show_bug\.cgi\?id=(?P<bug_id>\d+)(&ctype=xml)?"
    bug_url_short = r"http\://webkit\.org/b/(?P<bug_id>\d+)"

    def quips(self):
        # We only fetch and parse the list of quips once per instantiation
        # so that we do not burden bugs.webkit.org.
        if not self.cached_quips and not self.dryrun:
            self.cached_quips = self.queries.fetch_quips()
        return self.cached_quips

    def bug_url_for_bug_id(self, bug_id, xml=False):
        if not bug_id:
            return None
        content_type = "&ctype=xml" if xml else ""
        return "%sshow_bug.cgi?id=%s%s" % (self.bug_server_url, bug_id, content_type)

    def short_bug_url_for_bug_id(self, bug_id):
        if not bug_id:
            return None
        return "http://webkit.org/b/%s" % bug_id

    def add_attachment_url(self, bug_id):
        return "%sattachment.cgi?action=enter&bugid=%s" % (self.bug_server_url, bug_id)

    def attachment_url_for_id(self, attachment_id, action="view"):
        if not attachment_id:
            return None
        action_param = ""
        if action and action != "view":
            action_param = "&action=%s" % action
        return "%sattachment.cgi?id=%s%s" % (self.bug_server_url,
                                             attachment_id,
                                             action_param)

    def _parse_attachment_flag(self,
                               element,
                               flag_name,
                               attachment,
                               result_key):
        flag = element.find('flag', attrs={'name': flag_name})
        if flag:
            attachment[flag_name] = flag['status']
            if flag['status'] == '+':
                attachment[result_key] = flag['setter']
        # Sadly show_bug.cgi?ctype=xml does not expose the flag modification date.

    def _string_contents(self, soup):
        # WebKit's bugzilla instance uses UTF-8.
        # BeautifulSoup always returns Unicode strings, however
        # the .string method returns a (unicode) NavigableString.
        # NavigableString can confuse other parts of the code, so we
        # convert from NavigableString to a real unicode() object using unicode().
        return unicode(soup.string)

    # Example: 2010-01-20 14:31 PST
    # FIXME: Some bugzilla dates seem to have seconds in them?
    # Python does not support timezones out of the box.
    # Assume that bugzilla always uses PST (which is true for bugs.webkit.org)
    _bugzilla_date_format = "%Y-%m-%d %H:%M"

    @classmethod
    def _parse_date(cls, date_string):
        (date, time, time_zone) = date_string.split(" ")
        # Ignore the timezone because python doesn't understand timezones out of the box.
        date_string = "%s %s" % (date, time)
        return datetime.strptime(date_string, cls._bugzilla_date_format)

    def _date_contents(self, soup):
        return self._parse_date(self._string_contents(soup))

    def _parse_attachment_element(self, element, bug_id):
        attachment = {}
        attachment['bug_id'] = bug_id
        attachment['is_obsolete'] = (element.has_key('isobsolete') and element['isobsolete'] == "1")
        attachment['is_patch'] = (element.has_key('ispatch') and element['ispatch'] == "1")
        attachment['id'] = int(element.find('attachid').string)
        # FIXME: No need to parse out the url here.
        attachment['url'] = self.attachment_url_for_id(attachment['id'])
        attachment["attach_date"] = self._date_contents(element.find("date"))
        attachment['name'] = self._string_contents(element.find('desc'))
        attachment['attacher_email'] = self._string_contents(element.find('attacher'))
        attachment['type'] = self._string_contents(element.find('type'))
        self._parse_attachment_flag(
                element, 'review', attachment, 'reviewer_email')
        self._parse_attachment_flag(
                element, 'commit-queue', attachment, 'committer_email')
        return attachment

    def _parse_bugs_from_xml(self, page):
        soup = BeautifulSoup(page)
        # Without the unicode() call, BeautifulSoup occasionally complains of being
        # passed None for no apparent reason.
        return [Bug(self._parse_bug_dictionary_from_xml(unicode(bug_xml)), self) for bug_xml in soup('bug')]

    def _parse_bug_dictionary_from_xml(self, page):
        soup = BeautifulSoup(page)
        bug = {}
        bug["id"] = int(soup.find("bug_id").string)
        bug["title"] = self._string_contents(soup.find("short_desc"))
        bug["bug_status"] = self._string_contents(soup.find("bug_status"))
        dup_id = soup.find("dup_id")
        if dup_id:
            bug["dup_id"] = self._string_contents(dup_id)
        bug["reporter_email"] = self._string_contents(soup.find("reporter"))
        bug["assigned_to_email"] = self._string_contents(soup.find("assigned_to"))
        bug["cc_emails"] = [self._string_contents(element) for element in soup.findAll('cc')]
        bug["attachments"] = [self._parse_attachment_element(element, bug["id"]) for element in soup.findAll('attachment')]
        return bug

    # Makes testing fetch_*_from_bug() possible until we have a better
    # BugzillaNetwork abstration.

    def _fetch_bug_page(self, bug_id):
        bug_url = self.bug_url_for_bug_id(bug_id, xml=True)
        log("Fetching: %s" % bug_url)
        return self.browser.open(bug_url)

    def fetch_bug_dictionary(self, bug_id):
        try:
            return self._parse_bug_dictionary_from_xml(self._fetch_bug_page(bug_id))
        except KeyboardInterrupt:
            raise
        except:
            self.authenticate()
            return self._parse_bug_dictionary_from_xml(self._fetch_bug_page(bug_id))

    # FIXME: A BugzillaCache object should provide all these fetch_ methods.

    def fetch_bug(self, bug_id):
        return Bug(self.fetch_bug_dictionary(bug_id), self)

    def fetch_attachment_contents(self, attachment_id):
        attachment_url = self.attachment_url_for_id(attachment_id)
        # We need to authenticate to download patches from security bugs.
        self.authenticate()
        return self.browser.open(attachment_url).read()

    def _parse_bug_id_from_attachment_page(self, page):
        # The "Up" relation happens to point to the bug.
        up_link = BeautifulSoup(page).find('link', rel='Up')
        if not up_link:
            # This attachment does not exist (or you don't have permissions to
            # view it).
            return None
        match = re.search("show_bug.cgi\?id=(?P<bug_id>\d+)", up_link['href'])
        return int(match.group('bug_id'))

    def bug_id_for_attachment_id(self, attachment_id):
        self.authenticate()

        attachment_url = self.attachment_url_for_id(attachment_id, 'edit')
        log("Fetching: %s" % attachment_url)
        page = self.browser.open(attachment_url)
        return self._parse_bug_id_from_attachment_page(page)

    # FIXME: This should just return Attachment(id), which should be able to
    # lazily fetch needed data.

    def fetch_attachment(self, attachment_id):
        # We could grab all the attachment details off of the attachment edit
        # page but we already have working code to do so off of the bugs page,
        # so re-use that.
        bug_id = self.bug_id_for_attachment_id(attachment_id)
        if not bug_id:
            return None
        attachments = self.fetch_bug(bug_id).attachments(include_obsolete=True)
        for attachment in attachments:
            if attachment.id() == int(attachment_id):
                return attachment
        return None # This should never be hit.

    def authenticate(self):
        if self.authenticated:
            return

        if self.dryrun:
            log("Skipping log in for dry run...")
            self.authenticated = True
            return

        credentials = Credentials(self.bug_server_host, git_prefix="bugzilla")

        attempts = 0
        while not self.authenticated:
            attempts += 1
            username, password = credentials.read_credentials()

            log("Logging in as %s..." % username)
            self.browser.open(self.bug_server_url +
                              "index.cgi?GoAheadAndLogIn=1")
            self.browser.select_form(name="login")
            self.browser['Bugzilla_login'] = username
            self.browser['Bugzilla_password'] = password
            response = self.browser.submit()

            match = re.search("<title>(.+?)</title>", response.read())
            # If the resulting page has a title, and it contains the word
            # "invalid" assume it's the login failure page.
            if match and re.search("Invalid", match.group(1), re.IGNORECASE):
                errorMessage = "Bugzilla login failed: %s" % match.group(1)
                # raise an exception only if this was the last attempt
                if attempts < 5:
                    log(errorMessage)
                else:
                    raise Exception(errorMessage)
            else:
                self.authenticated = True
                self.username = username

    def _commit_queue_flag(self, mark_for_landing, mark_for_commit_queue):
        if mark_for_landing:
            return '+'
        elif mark_for_commit_queue:
            return '?'
        return 'X'

    # FIXME: mark_for_commit_queue and mark_for_landing should be joined into a single commit_flag argument.
    def _fill_attachment_form(self,
                              description,
                              file_object,
                              mark_for_review=False,
                              mark_for_commit_queue=False,
                              mark_for_landing=False,
                              is_patch=False,
                              filename=None,
                              mimetype=None):
        self.browser['description'] = description
        if is_patch:
            self.browser['ispatch'] = ("1",)
        # FIXME: Should this use self._find_select_element_for_flag?
        self.browser['flag_type-1'] = ('?',) if mark_for_review else ('X',)
        self.browser['flag_type-3'] = (self._commit_queue_flag(mark_for_landing, mark_for_commit_queue),)

        filename = filename or "%s.patch" % timestamp()
        if not mimetype:
            mimetypes.add_type('text/plain', '.patch')  # Make sure mimetypes knows about .patch
            mimetype, _ = mimetypes.guess_type(filename)
        if not mimetype:
            mimetype = "text/plain"  # Bugzilla might auto-guess for us and we might not need this?
        self.browser.add_file(file_object, mimetype, filename, 'data')

    def _file_object_for_upload(self, file_or_string):
        if hasattr(file_or_string, 'read'):
            return file_or_string
        # Only if file_or_string is not already encoded do we want to encode it.
        if isinstance(file_or_string, unicode):
            file_or_string = file_or_string.encode('utf-8')
        return StringIO.StringIO(file_or_string)

    # timestamp argument is just for unittests.
    def _filename_for_upload(self, file_object, bug_id, extension="txt", timestamp=timestamp):
        if hasattr(file_object, "name"):
            return file_object.name
        return "bug-%s-%s.%s" % (bug_id, timestamp(), extension)

    def add_attachment_to_bug(self,
                              bug_id,
                              file_or_string,
                              description,
                              filename=None,
                              comment_text=None):
        self.authenticate()
        log('Adding attachment "%s" to %s' % (description, self.bug_url_for_bug_id(bug_id)))
        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.add_attachment_url(bug_id))
        self.browser.select_form(name="entryform")
        file_object = self._file_object_for_upload(file_or_string)
        filename = filename or self._filename_for_upload(file_object, bug_id)
        self._fill_attachment_form(description, file_object, filename=filename)
        if comment_text:
            log(comment_text)
            self.browser['comment'] = comment_text
        self.browser.submit()

    # FIXME: The arguments to this function should be simplified and then
    # this should be merged into add_attachment_to_bug
    def add_patch_to_bug(self,
                         bug_id,
                         file_or_string,
                         description,
                         comment_text=None,
                         mark_for_review=False,
                         mark_for_commit_queue=False,
                         mark_for_landing=False):
        self.authenticate()
        log('Adding patch "%s" to %s' % (description, self.bug_url_for_bug_id(bug_id)))

        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.add_attachment_url(bug_id))
        self.browser.select_form(name="entryform")
        file_object = self._file_object_for_upload(file_or_string)
        filename = self._filename_for_upload(file_object, bug_id, extension="patch")
        self._fill_attachment_form(description,
                                   file_object,
                                   mark_for_review=mark_for_review,
                                   mark_for_commit_queue=mark_for_commit_queue,
                                   mark_for_landing=mark_for_landing,
                                   is_patch=True,
                                   filename=filename)
        if comment_text:
            log(comment_text)
            self.browser['comment'] = comment_text
        self.browser.submit()

    # FIXME: There has to be a more concise way to write this method.
    def _check_create_bug_response(self, response_html):
        match = re.search("<title>Bug (?P<bug_id>\d+) Submitted</title>",
                          response_html)
        if match:
            return match.group('bug_id')

        match = re.search(
            '<div id="bugzilla-body">(?P<error_message>.+)<div id="footer">',
            response_html,
            re.DOTALL)
        error_message = "FAIL"
        if match:
            text_lines = BeautifulSoup(
                    match.group('error_message')).findAll(text=True)
            error_message = "\n" + '\n'.join(
                    ["  " + line.strip()
                     for line in text_lines if line.strip()])
        raise Exception("Bug not created: %s" % error_message)

    def create_bug(self,
                   bug_title,
                   bug_description,
                   component=None,
                   diff=None,
                   patch_description=None,
                   cc=None,
                   blocked=None,
                   assignee=None,
                   mark_for_review=False,
                   mark_for_commit_queue=False):
        self.authenticate()

        log('Creating bug with title "%s"' % bug_title)
        if self.dryrun:
            log(bug_description)
            # FIXME: This will make some paths fail, as they assume this returns an id.
            return

        self.browser.open(self.bug_server_url + "enter_bug.cgi?product=WebKit")
        self.browser.select_form(name="Create")
        component_items = self.browser.find_control('component').items
        component_names = map(lambda item: item.name, component_items)
        if not component:
            component = "New Bugs"
        if component not in component_names:
            component = User.prompt_with_list("Please pick a component:", component_names)
        self.browser["component"] = [component]
        if cc:
            self.browser["cc"] = cc
        if blocked:
            self.browser["blocked"] = unicode(blocked)
        if not assignee:
            assignee = self.username
        if assignee and not self.browser.find_control("assigned_to").disabled:
            self.browser["assigned_to"] = assignee
        self.browser["short_desc"] = bug_title
        self.browser["comment"] = bug_description

        if diff:
            # _fill_attachment_form expects a file-like object
            # Patch files are already binary, so no encoding needed.
            assert(isinstance(diff, str))
            patch_file_object = StringIO.StringIO(diff)
            self._fill_attachment_form(
                    patch_description,
                    patch_file_object,
                    mark_for_review=mark_for_review,
                    mark_for_commit_queue=mark_for_commit_queue,
                    is_patch=True)

        response = self.browser.submit()

        bug_id = self._check_create_bug_response(response.read())
        log("Bug %s created." % bug_id)
        log("%sshow_bug.cgi?id=%s" % (self.bug_server_url, bug_id))
        return bug_id

    def _find_select_element_for_flag(self, flag_name):
        # FIXME: This will break if we ever re-order attachment flags
        if flag_name == "review":
            return self.browser.find_control(type='select', nr=0)
        elif flag_name == "commit-queue":
            return self.browser.find_control(type='select', nr=1)
        raise Exception("Don't know how to find flag named \"%s\"" % flag_name)

    def clear_attachment_flags(self,
                               attachment_id,
                               additional_comment_text=None):
        self.authenticate()

        comment_text = "Clearing flags on attachment: %s" % attachment_id
        if additional_comment_text:
            comment_text += "\n\n%s" % additional_comment_text
        log(comment_text)

        if self.dryrun:
            return

        self.browser.open(self.attachment_url_for_id(attachment_id, 'edit'))
        self.browser.select_form(nr=1)
        self.browser.set_value(comment_text, name='comment', nr=0)
        self._find_select_element_for_flag('review').value = ("X",)
        self._find_select_element_for_flag('commit-queue').value = ("X",)
        self.browser.submit()

    def set_flag_on_attachment(self,
                               attachment_id,
                               flag_name,
                               flag_value,
                               comment_text=None,
                               additional_comment_text=None):
        # FIXME: We need a way to test this function on a live bugzilla
        # instance.

        self.authenticate()

        if additional_comment_text:
            comment_text += "\n\n%s" % additional_comment_text
        log(comment_text)

        if self.dryrun:
            return

        self.browser.open(self.attachment_url_for_id(attachment_id, 'edit'))
        self.browser.select_form(nr=1)

        if comment_text:
            self.browser.set_value(comment_text, name='comment', nr=0)

        self._find_select_element_for_flag(flag_name).value = (flag_value,)
        self.browser.submit()

    # FIXME: All of these bug editing methods have a ridiculous amount of
    # copy/paste code.

    def obsolete_attachment(self, attachment_id, comment_text=None):
        self.authenticate()

        log("Obsoleting attachment: %s" % attachment_id)
        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.attachment_url_for_id(attachment_id, 'edit'))
        self.browser.select_form(nr=1)
        self.browser.find_control('isobsolete').items[0].selected = True
        # Also clear any review flag (to remove it from review/commit queues)
        self._find_select_element_for_flag('review').value = ("X",)
        self._find_select_element_for_flag('commit-queue').value = ("X",)
        if comment_text:
            log(comment_text)
            # Bugzilla has two textareas named 'comment', one is somehow
            # hidden.  We want the first.
            self.browser.set_value(comment_text, name='comment', nr=0)
        self.browser.submit()

    def add_cc_to_bug(self, bug_id, email_address_list):
        self.authenticate()

        log("Adding %s to the CC list for bug %s" % (email_address_list,
                                                     bug_id))
        if self.dryrun:
            return

        self.browser.open(self.bug_url_for_bug_id(bug_id))
        self.browser.select_form(name="changeform")
        self.browser["newcc"] = ", ".join(email_address_list)
        self.browser.submit()

    def post_comment_to_bug(self, bug_id, comment_text, cc=None):
        self.authenticate()

        log("Adding comment to bug %s" % bug_id)
        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.bug_url_for_bug_id(bug_id))
        self.browser.select_form(name="changeform")
        self.browser["comment"] = comment_text
        if cc:
            self.browser["newcc"] = ", ".join(cc)
        self.browser.submit()

    def close_bug_as_fixed(self, bug_id, comment_text=None):
        self.authenticate()

        log("Closing bug %s as fixed" % bug_id)
        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.bug_url_for_bug_id(bug_id))
        self.browser.select_form(name="changeform")
        if comment_text:
            self.browser['comment'] = comment_text
        self.browser['bug_status'] = ['RESOLVED']
        self.browser['resolution'] = ['FIXED']
        self.browser.submit()

    def reassign_bug(self, bug_id, assignee, comment_text=None):
        self.authenticate()

        log("Assigning bug %s to %s" % (bug_id, assignee))
        if self.dryrun:
            log(comment_text)
            return

        self.browser.open(self.bug_url_for_bug_id(bug_id))
        self.browser.select_form(name="changeform")
        if comment_text:
            log(comment_text)
            self.browser["comment"] = comment_text
        self.browser["assigned_to"] = assignee
        self.browser.submit()

    def reopen_bug(self, bug_id, comment_text):
        self.authenticate()

        log("Re-opening bug %s" % bug_id)
        # Bugzilla requires a comment when re-opening a bug, so we know it will
        # never be None.
        log(comment_text)
        if self.dryrun:
            return

        self.browser.open(self.bug_url_for_bug_id(bug_id))
        self.browser.select_form(name="changeform")
        bug_status = self.browser.find_control("bug_status", type="select")
        # This is a hack around the fact that ClientForm.ListControl seems to
        # have no simpler way to ask if a control has an item named "REOPENED"
        # without using exceptions for control flow.
        possible_bug_statuses = map(lambda item: item.name, bug_status.items)
        if "REOPENED" in possible_bug_statuses:
            bug_status.value = ["REOPENED"]
        # If the bug was never confirmed it will not have a "REOPENED"
        # state, but only an "UNCONFIRMED" state.
        elif "UNCONFIRMED" in possible_bug_statuses:
            bug_status.value = ["UNCONFIRMED"]
        else:
            # FIXME: This logic is slightly backwards.  We won't print this
            # message if the bug is already open with state "UNCONFIRMED".
            log("Did not reopen bug %s, it appears to already be open with status %s." % (bug_id, bug_status.value))
        self.browser['comment'] = comment_text
        self.browser.submit()