1
# Copyright (C) 2005, 2006, 2007, 2008, 2009 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
# TODO: Perhaps there should be an API to find out if bzr running under the
19
# test suite -- some plugins might want to avoid making intrusive changes if
20
# this is the case. However, we want behaviour under to test to diverge as
21
# little as possible, so this should be used rarely if it's added at all.
22
# (Suggestion from j-a-meinel, 2005-11-24)
24
# NOTE: Some classes in here use camelCaseNaming() rather than
25
# underscore_naming(). That's for consistency with unittest; it's not the
26
# general style of bzrlib. Please continue that consistency when adding e.g.
27
# new assertFoo() methods.
31
from cStringIO import StringIO
38
from pprint import pformat
43
from subprocess import Popen, PIPE, STDOUT
68
import bzrlib.commands
69
import bzrlib.timestamp
71
import bzrlib.inventory
72
import bzrlib.iterablefile
77
# lsprof not available
79
from bzrlib.merge import merge_inner
82
from bzrlib.smart import client, request, server
84
from bzrlib import symbol_versioning
85
from bzrlib.symbol_versioning import (
92
from bzrlib.transport import get_transport
93
import bzrlib.transport
94
from bzrlib.transport.local import LocalURLServer
95
from bzrlib.transport.memory import MemoryServer
96
from bzrlib.transport.readonly import ReadonlyServer
97
from bzrlib.trace import mutter, note
98
from bzrlib.tests import TestUtil
99
from bzrlib.tests.http_server import HttpServer
100
from bzrlib.tests.TestUtil import (
104
from bzrlib.tests.treeshape import build_tree_contents
105
from bzrlib.ui.text import TextUIFactory
106
import bzrlib.version_info_formats.format_custom
107
from bzrlib.workingtree import WorkingTree, WorkingTreeFormat2
109
# Mark this python module as being part of the implementation
110
# of unittest: this gives us better tracebacks where the last
111
# shown frame is the test code, not our assertXYZ.
114
default_transport = LocalURLServer
117
class ExtendedTestResult(unittest._TextTestResult):
118
"""Accepts, reports and accumulates the results of running tests.
120
Compared to the unittest version this class adds support for
121
profiling, benchmarking, stopping as soon as a test fails, and
122
skipping tests. There are further-specialized subclasses for
123
different types of display.
125
When a test finishes, in whatever way, it calls one of the addSuccess,
126
addFailure or addError classes. These in turn may redirect to a more
127
specific case for the special test results supported by our extended
130
Note that just one of these objects is fed the results from many tests.
135
def __init__(self, stream, descriptions, verbosity,
140
"""Construct new TestResult.
142
:param bench_history: Optionally, a writable file object to accumulate
145
unittest._TextTestResult.__init__(self, stream, descriptions, verbosity)
146
if bench_history is not None:
147
from bzrlib.version import _get_bzr_source_tree
148
src_tree = _get_bzr_source_tree()
151
revision_id = src_tree.get_parent_ids()[0]
153
# XXX: if this is a brand new tree, do the same as if there
157
# XXX: If there's no branch, what should we do?
159
bench_history.write("--date %s %s\n" % (time.time(), revision_id))
160
self._bench_history = bench_history
161
self.ui = ui.ui_factory
162
self.num_tests = num_tests
164
self.failure_count = 0
165
self.known_failure_count = 0
167
self.not_applicable_count = 0
168
self.unsupported = {}
170
self._overall_start_time = time.time()
171
self._strict = strict
175
ok = self.wasStrictlySuccessful()
177
ok = self.wasSuccessful()
179
self.stream.write('tests passed\n')
181
self.stream.write('tests failed\n')
182
if TestCase._first_thread_leaker_id:
184
'%s is leaking threads among %d leaking tests.\n' % (
185
TestCase._first_thread_leaker_id,
186
TestCase._leaking_threads_tests))
188
def _extractBenchmarkTime(self, testCase):
189
"""Add a benchmark time for the current test case."""
190
return getattr(testCase, "_benchtime", None)
192
def _elapsedTestTimeString(self):
193
"""Return a time string for the overall time the current test has taken."""
194
return self._formatTime(time.time() - self._start_time)
196
def _testTimeString(self, testCase):
197
benchmark_time = self._extractBenchmarkTime(testCase)
198
if benchmark_time is not None:
199
return self._formatTime(benchmark_time) + "*"
201
return self._elapsedTestTimeString()
203
def _formatTime(self, seconds):
204
"""Format seconds as milliseconds with leading spaces."""
205
# some benchmarks can take thousands of seconds to run, so we need 8
207
return "%8dms" % (1000 * seconds)
209
def _shortened_test_description(self, test):
211
what = re.sub(r'^bzrlib\.(tests|benchmarks)\.', '', what)
214
def startTest(self, test):
215
unittest.TestResult.startTest(self, test)
218
self.report_test_start(test)
219
test.number = self.count
220
self._recordTestStartTime()
222
def startTests(self):
224
'testing: %s\n' % (osutils.realpath(sys.argv[0]),))
226
' %s (%s python%s)\n' % (
228
bzrlib.version_string,
229
bzrlib._format_version_tuple(sys.version_info),
231
self.stream.write('\n')
233
def _recordTestStartTime(self):
234
"""Record that a test has started."""
235
self._start_time = time.time()
237
def _cleanupLogFile(self, test):
238
# We can only do this if we have one of our TestCases, not if
240
setKeepLogfile = getattr(test, 'setKeepLogfile', None)
241
if setKeepLogfile is not None:
244
def addError(self, test, err):
245
"""Tell result that test finished with an error.
247
Called from the TestCase run() method when the test
248
fails with an unexpected error.
250
self._testConcluded(test)
251
if isinstance(err[1], TestNotApplicable):
252
return self._addNotApplicable(test, err)
253
elif isinstance(err[1], UnavailableFeature):
254
return self.addNotSupported(test, err[1].args[0])
256
unittest.TestResult.addError(self, test, err)
257
self.error_count += 1
258
self.report_error(test, err)
261
self._cleanupLogFile(test)
263
def addFailure(self, test, err):
264
"""Tell result that test failed.
266
Called from the TestCase run() method when the test
267
fails because e.g. an assert() method failed.
269
self._testConcluded(test)
270
if isinstance(err[1], KnownFailure):
271
return self._addKnownFailure(test, err)
273
unittest.TestResult.addFailure(self, test, err)
274
self.failure_count += 1
275
self.report_failure(test, err)
278
self._cleanupLogFile(test)
280
def addSuccess(self, test):
281
"""Tell result that test completed successfully.
283
Called from the TestCase run()
285
self._testConcluded(test)
286
if self._bench_history is not None:
287
benchmark_time = self._extractBenchmarkTime(test)
288
if benchmark_time is not None:
289
self._bench_history.write("%s %s\n" % (
290
self._formatTime(benchmark_time),
292
self.report_success(test)
293
self._cleanupLogFile(test)
294
unittest.TestResult.addSuccess(self, test)
295
test._log_contents = ''
297
def _testConcluded(self, test):
298
"""Common code when a test has finished.
300
Called regardless of whether it succeded, failed, etc.
304
def _addKnownFailure(self, test, err):
305
self.known_failure_count += 1
306
self.report_known_failure(test, err)
308
def addNotSupported(self, test, feature):
309
"""The test will not be run because of a missing feature.
311
# this can be called in two different ways: it may be that the
312
# test started running, and then raised (through addError)
313
# UnavailableFeature. Alternatively this method can be called
314
# while probing for features before running the tests; in that
315
# case we will see startTest and stopTest, but the test will never
317
self.unsupported.setdefault(str(feature), 0)
318
self.unsupported[str(feature)] += 1
319
self.report_unsupported(test, feature)
321
def addSkip(self, test, reason):
322
"""A test has not run for 'reason'."""
324
self.report_skip(test, reason)
326
def _addNotApplicable(self, test, skip_excinfo):
327
if isinstance(skip_excinfo[1], TestNotApplicable):
328
self.not_applicable_count += 1
329
self.report_not_applicable(test, skip_excinfo)
332
except KeyboardInterrupt:
335
self.addError(test, test.exc_info())
337
# seems best to treat this as success from point-of-view of unittest
338
# -- it actually does nothing so it barely matters :)
339
unittest.TestResult.addSuccess(self, test)
340
test._log_contents = ''
342
def printErrorList(self, flavour, errors):
343
for test, err in errors:
344
self.stream.writeln(self.separator1)
345
self.stream.write("%s: " % flavour)
346
self.stream.writeln(self.getDescription(test))
347
if getattr(test, '_get_log', None) is not None:
348
log_contents = test._get_log()
350
self.stream.write('\n')
352
('vvvv[log from %s]' % test.id()).ljust(78,'-'))
353
self.stream.write('\n')
354
self.stream.write(log_contents)
355
self.stream.write('\n')
357
('^^^^[log from %s]' % test.id()).ljust(78,'-'))
358
self.stream.write('\n')
359
self.stream.writeln(self.separator2)
360
self.stream.writeln("%s" % err)
365
def report_cleaning_up(self):
368
def report_success(self, test):
371
def wasStrictlySuccessful(self):
372
if self.unsupported or self.known_failure_count:
374
return self.wasSuccessful()
377
class TextTestResult(ExtendedTestResult):
378
"""Displays progress and results of tests in text form"""
380
def __init__(self, stream, descriptions, verbosity,
386
ExtendedTestResult.__init__(self, stream, descriptions, verbosity,
387
bench_history, num_tests, strict)
388
# We no longer pass them around, but just rely on the UIFactory stack
391
warnings.warn("Passing pb to TextTestResult is deprecated")
392
self.pb = self.ui.nested_progress_bar()
393
self.pb.show_pct = False
394
self.pb.show_spinner = False
395
self.pb.show_eta = False,
396
self.pb.show_count = False
397
self.pb.show_bar = False
398
self.pb.update_latency = 0
399
self.pb.show_transport_activity = False
401
def report_starting(self):
402
self.pb.update('[test 0/%d] Starting' % (self.num_tests))
404
def _progress_prefix_text(self):
405
# the longer this text, the less space we have to show the test
407
a = '[%d' % self.count # total that have been run
408
# tests skipped as known not to be relevant are not important enough
410
## if self.skip_count:
411
## a += ', %d skip' % self.skip_count
412
## if self.known_failure_count:
413
## a += '+%dX' % self.known_failure_count
414
if self.num_tests is not None:
415
a +='/%d' % self.num_tests
417
runtime = time.time() - self._overall_start_time
419
a += '%dm%ds' % (runtime / 60, runtime % 60)
423
a += ', %d err' % self.error_count
424
if self.failure_count:
425
a += ', %d fail' % self.failure_count
427
a += ', %d missing' % len(self.unsupported)
431
def report_test_start(self, test):
434
self._progress_prefix_text()
436
+ self._shortened_test_description(test))
438
def _test_description(self, test):
439
return self._shortened_test_description(test)
441
def report_error(self, test, err):
442
self.pb.note('ERROR: %s\n %s\n',
443
self._test_description(test),
447
def report_failure(self, test, err):
448
self.pb.note('FAIL: %s\n %s\n',
449
self._test_description(test),
453
def report_known_failure(self, test, err):
454
self.pb.note('XFAIL: %s\n%s\n',
455
self._test_description(test), err[1])
457
def report_skip(self, test, reason):
460
def report_not_applicable(self, test, skip_excinfo):
463
def report_unsupported(self, test, feature):
464
"""test cannot be run because feature is missing."""
466
def report_cleaning_up(self):
467
self.pb.update('Cleaning up')
469
def printErrors(self):
470
# clear the pb to make room for the error listing
472
super(TextTestResult, self).printErrors()
478
class VerboseTestResult(ExtendedTestResult):
479
"""Produce long output, with one line per test run plus times"""
481
def _ellipsize_to_right(self, a_string, final_width):
482
"""Truncate and pad a string, keeping the right hand side"""
483
if len(a_string) > final_width:
484
result = '...' + a_string[3-final_width:]
487
return result.ljust(final_width)
489
def report_starting(self):
490
self.stream.write('running %d tests...\n' % self.num_tests)
492
def report_test_start(self, test):
494
name = self._shortened_test_description(test)
495
# width needs space for 6 char status, plus 1 for slash, plus an
496
# 11-char time string, plus a trailing blank
497
# when NUMBERED_DIRS: plus 5 chars on test number, plus 1 char on space
498
self.stream.write(self._ellipsize_to_right(name,
499
osutils.terminal_width()-18))
502
def _error_summary(self, err):
504
return '%s%s' % (indent, err[1])
506
def report_error(self, test, err):
507
self.stream.writeln('ERROR %s\n%s'
508
% (self._testTimeString(test),
509
self._error_summary(err)))
511
def report_failure(self, test, err):
512
self.stream.writeln(' FAIL %s\n%s'
513
% (self._testTimeString(test),
514
self._error_summary(err)))
516
def report_known_failure(self, test, err):
517
self.stream.writeln('XFAIL %s\n%s'
518
% (self._testTimeString(test),
519
self._error_summary(err)))
521
def report_success(self, test):
522
self.stream.writeln(' OK %s' % self._testTimeString(test))
523
for bench_called, stats in getattr(test, '_benchcalls', []):
524
self.stream.writeln('LSProf output for %s(%s, %s)' % bench_called)
525
stats.pprint(file=self.stream)
526
# flush the stream so that we get smooth output. This verbose mode is
527
# used to show the output in PQM.
530
def report_skip(self, test, reason):
531
self.stream.writeln(' SKIP %s\n%s'
532
% (self._testTimeString(test), reason))
534
def report_not_applicable(self, test, skip_excinfo):
535
self.stream.writeln(' N/A %s\n%s'
536
% (self._testTimeString(test),
537
self._error_summary(skip_excinfo)))
539
def report_unsupported(self, test, feature):
540
"""test cannot be run because feature is missing."""
541
self.stream.writeln("NODEP %s\n The feature '%s' is not available."
542
%(self._testTimeString(test), feature))
545
class TextTestRunner(object):
546
stop_on_failure = False
556
self.stream = unittest._WritelnDecorator(stream)
557
self.descriptions = descriptions
558
self.verbosity = verbosity
559
self._bench_history = bench_history
560
self.list_only = list_only
561
self._strict = strict
564
"Run the given test case or test suite."
565
startTime = time.time()
566
if self.verbosity == 1:
567
result_class = TextTestResult
568
elif self.verbosity >= 2:
569
result_class = VerboseTestResult
570
result = result_class(self.stream,
573
bench_history=self._bench_history,
574
num_tests=test.countTestCases(),
577
result.stop_early = self.stop_on_failure
578
result.report_starting()
580
if self.verbosity >= 2:
581
self.stream.writeln("Listing tests only ...\n")
583
for t in iter_suite_tests(test):
584
self.stream.writeln("%s" % (t.id()))
593
if isinstance(test, testtools.ConcurrentTestSuite):
594
# We need to catch bzr specific behaviors
595
test.run(BZRTransformingResult(result))
598
run = result.testsRun
600
stopTime = time.time()
601
timeTaken = stopTime - startTime
603
self.stream.writeln(result.separator2)
604
self.stream.writeln("%s %d test%s in %.3fs" % (actionTaken,
605
run, run != 1 and "s" or "", timeTaken))
606
self.stream.writeln()
607
if not result.wasSuccessful():
608
self.stream.write("FAILED (")
609
failed, errored = map(len, (result.failures, result.errors))
611
self.stream.write("failures=%d" % failed)
613
if failed: self.stream.write(", ")
614
self.stream.write("errors=%d" % errored)
615
if result.known_failure_count:
616
if failed or errored: self.stream.write(", ")
617
self.stream.write("known_failure_count=%d" %
618
result.known_failure_count)
619
self.stream.writeln(")")
621
if result.known_failure_count:
622
self.stream.writeln("OK (known_failures=%d)" %
623
result.known_failure_count)
625
self.stream.writeln("OK")
626
if result.skip_count > 0:
627
skipped = result.skip_count
628
self.stream.writeln('%d test%s skipped' %
629
(skipped, skipped != 1 and "s" or ""))
630
if result.unsupported:
631
for feature, count in sorted(result.unsupported.items()):
632
self.stream.writeln("Missing feature '%s' skipped %d tests." %
638
def iter_suite_tests(suite):
639
"""Return all tests in a suite, recursing through nested suites"""
640
if isinstance(suite, unittest.TestCase):
642
elif isinstance(suite, unittest.TestSuite):
644
for r in iter_suite_tests(item):
647
raise Exception('unknown type %r for object %r'
648
% (type(suite), suite))
651
class TestSkipped(Exception):
652
"""Indicates that a test was intentionally skipped, rather than failing."""
655
class TestNotApplicable(TestSkipped):
656
"""A test is not applicable to the situation where it was run.
658
This is only normally raised by parameterized tests, if they find that
659
the instance they're constructed upon does not support one aspect
664
class KnownFailure(AssertionError):
665
"""Indicates that a test failed in a precisely expected manner.
667
Such failures dont block the whole test suite from passing because they are
668
indicators of partially completed code or of future work. We have an
669
explicit error for them so that we can ensure that they are always visible:
670
KnownFailures are always shown in the output of bzr selftest.
674
class UnavailableFeature(Exception):
675
"""A feature required for this test was not available.
677
The feature should be used to construct the exception.
681
class CommandFailed(Exception):
685
class StringIOWrapper(object):
686
"""A wrapper around cStringIO which just adds an encoding attribute.
688
Internally we can check sys.stdout to see what the output encoding
689
should be. However, cStringIO has no encoding attribute that we can
690
set. So we wrap it instead.
695
def __init__(self, s=None):
697
self.__dict__['_cstring'] = StringIO(s)
699
self.__dict__['_cstring'] = StringIO()
701
def __getattr__(self, name, getattr=getattr):
702
return getattr(self.__dict__['_cstring'], name)
704
def __setattr__(self, name, val):
705
if name == 'encoding':
706
self.__dict__['encoding'] = val
708
return setattr(self._cstring, name, val)
711
class TestUIFactory(TextUIFactory):
712
"""A UI Factory for testing.
714
Hide the progress bar but emit note()s.
716
Allows get_password to be tested without real tty attached.
719
def __init__(self, stdout=None, stderr=None, stdin=None):
720
if stdin is not None:
721
# We use a StringIOWrapper to be able to test various
722
# encodings, but the user is still responsible to
723
# encode the string and to set the encoding attribute
724
# of StringIOWrapper.
725
stdin = StringIOWrapper(stdin)
726
super(TestUIFactory, self).__init__(stdin, stdout, stderr)
729
"""See progress.ProgressBar.clear()."""
731
def clear_term(self):
732
"""See progress.ProgressBar.clear_term()."""
735
"""See progress.ProgressBar.finished()."""
737
def note(self, fmt_string, *args):
738
"""See progress.ProgressBar.note()."""
740
fmt_string = fmt_string % args
741
self.stdout.write(fmt_string + "\n")
743
def progress_bar(self):
746
def nested_progress_bar(self):
749
def update(self, message, count=None, total=None):
750
"""See progress.ProgressBar.update()."""
752
def get_non_echoed_password(self):
753
"""Get password from stdin without trying to handle the echo mode"""
754
password = self.stdin.readline()
757
if password[-1] == '\n':
758
password = password[:-1]
762
class TestCase(unittest.TestCase):
763
"""Base class for bzr unit tests.
765
Tests that need access to disk resources should subclass
766
TestCaseInTempDir not TestCase.
768
Error and debug log messages are redirected from their usual
769
location into a temporary file, the contents of which can be
770
retrieved by _get_log(). We use a real OS file, not an in-memory object,
771
so that it can also capture file IO. When the test completes this file
772
is read into memory and removed from disk.
774
There are also convenience functions to invoke bzr's command-line
775
routine, and to build and check bzr trees.
777
In addition to the usual method of overriding tearDown(), this class also
778
allows subclasses to register functions into the _cleanups list, which is
779
run in order as the object is torn down. It's less likely this will be
780
accidentally overlooked.
783
_active_threads = None
784
_leaking_threads_tests = 0
785
_first_thread_leaker_id = None
786
_log_file_name = None
788
_keep_log_file = False
789
# record lsprof data when performing benchmark calls.
790
_gather_lsprof_in_benchmarks = False
791
attrs_to_keep = ('id', '_testMethodName', '_testMethodDoc',
792
'_log_contents', '_log_file_name', '_benchtime',
793
'_TestCase__testMethodName', '_TestCase__testMethodDoc',)
795
def __init__(self, methodName='testMethod'):
796
super(TestCase, self).__init__(methodName)
798
self._bzr_test_setUp_run = False
799
self._bzr_test_tearDown_run = False
802
unittest.TestCase.setUp(self)
803
self._bzr_test_setUp_run = True
804
self._cleanEnvironment()
807
self._benchcalls = []
808
self._benchtime = None
810
# Track locks - needs to be called before _clear_debug_flags.
812
self._clear_debug_flags()
813
TestCase._active_threads = threading.activeCount()
814
self.addCleanup(self._check_leaked_threads)
819
pdb.Pdb().set_trace(sys._getframe().f_back)
821
def _check_leaked_threads(self):
822
active = threading.activeCount()
823
leaked_threads = active - TestCase._active_threads
824
TestCase._active_threads = active
826
TestCase._leaking_threads_tests += 1
827
if TestCase._first_thread_leaker_id is None:
828
TestCase._first_thread_leaker_id = self.id()
830
def _clear_debug_flags(self):
831
"""Prevent externally set debug flags affecting tests.
833
Tests that want to use debug flags can just set them in the
834
debug_flags set during setup/teardown.
836
self._preserved_debug_flags = set(debug.debug_flags)
837
if 'allow_debug' not in selftest_debug_flags:
838
debug.debug_flags.clear()
839
self.addCleanup(self._restore_debug_flags)
841
def _clear_hooks(self):
842
# prevent hooks affecting tests
843
self._preserved_hooks = {}
844
for key, factory in hooks.known_hooks.items():
845
parent, name = hooks.known_hooks_key_to_parent_and_attribute(key)
846
current_hooks = hooks.known_hooks_key_to_object(key)
847
self._preserved_hooks[parent] = (name, current_hooks)
848
self.addCleanup(self._restoreHooks)
849
for key, factory in hooks.known_hooks.items():
850
parent, name = hooks.known_hooks_key_to_parent_and_attribute(key)
851
setattr(parent, name, factory())
852
# this hook should always be installed
853
request._install_hook()
855
def _silenceUI(self):
856
"""Turn off UI for duration of test"""
857
# by default the UI is off; tests can turn it on if they want it.
858
saved = ui.ui_factory
860
ui.ui_factory = saved
861
ui.ui_factory = ui.SilentUIFactory()
862
self.addCleanup(_restore)
864
def _check_locks(self):
865
"""Check that all lock take/release actions have been paired."""
866
# once we have fixed all the current lock problems, we can change the
867
# following code to always check for mismatched locks, but only do
868
# traceback showing with -Dlock (self._lock_check_thorough is True).
869
# For now, because the test suite will fail, we only assert that lock
870
# matching has occured with -Dlock.
872
acquired_locks = [lock for action, lock in self._lock_actions
873
if action == 'acquired']
874
released_locks = [lock for action, lock in self._lock_actions
875
if action == 'released']
876
broken_locks = [lock for action, lock in self._lock_actions
877
if action == 'broken']
878
# trivially, given the tests for lock acquistion and release, if we
879
# have as many in each list, it should be ok. Some lock tests also
880
# break some locks on purpose and should be taken into account by
881
# considering that breaking a lock is just a dirty way of releasing it.
882
if len(acquired_locks) != (len(released_locks) + len(broken_locks)):
883
message = ('Different number of acquired and '
884
'released or broken locks. (%s, %s + %s)' %
885
(acquired_locks, released_locks, broken_locks))
886
if not self._lock_check_thorough:
887
# Rather than fail, just warn
888
print "Broken test %s: %s" % (self, message)
892
def _track_locks(self):
893
"""Track lock activity during tests."""
894
self._lock_actions = []
895
self._lock_check_thorough = 'lock' not in debug.debug_flags
896
self.addCleanup(self._check_locks)
897
_mod_lock.Lock.hooks.install_named_hook('lock_acquired',
898
self._lock_acquired, None)
899
_mod_lock.Lock.hooks.install_named_hook('lock_released',
900
self._lock_released, None)
901
_mod_lock.Lock.hooks.install_named_hook('lock_broken',
902
self._lock_broken, None)
904
def _lock_acquired(self, result):
905
self._lock_actions.append(('acquired', result))
907
def _lock_released(self, result):
908
self._lock_actions.append(('released', result))
910
def _lock_broken(self, result):
911
self._lock_actions.append(('broken', result))
913
def _ndiff_strings(self, a, b):
914
"""Return ndiff between two strings containing lines.
916
A trailing newline is added if missing to make the strings
918
if b and b[-1] != '\n':
920
if a and a[-1] != '\n':
922
difflines = difflib.ndiff(a.splitlines(True),
924
linejunk=lambda x: False,
925
charjunk=lambda x: False)
926
return ''.join(difflines)
928
def assertEqual(self, a, b, message=''):
932
except UnicodeError, e:
933
# If we can't compare without getting a UnicodeError, then
934
# obviously they are different
935
mutter('UnicodeError: %s', e)
938
raise AssertionError("%snot equal:\na = %s\nb = %s\n"
940
pformat(a), pformat(b)))
942
assertEquals = assertEqual
944
def assertEqualDiff(self, a, b, message=None):
945
"""Assert two texts are equal, if not raise an exception.
947
This is intended for use with multi-line strings where it can
948
be hard to find the differences by eye.
950
# TODO: perhaps override assertEquals to call this for strings?
954
message = "texts not equal:\n"
956
message = 'first string is missing a final newline.\n'
958
message = 'second string is missing a final newline.\n'
959
raise AssertionError(message +
960
self._ndiff_strings(a, b))
962
def assertEqualMode(self, mode, mode_test):
963
self.assertEqual(mode, mode_test,
964
'mode mismatch %o != %o' % (mode, mode_test))
966
def assertEqualStat(self, expected, actual):
967
"""assert that expected and actual are the same stat result.
969
:param expected: A stat result.
970
:param actual: A stat result.
971
:raises AssertionError: If the expected and actual stat values differ
974
self.assertEqual(expected.st_size, actual.st_size)
975
self.assertEqual(expected.st_mtime, actual.st_mtime)
976
self.assertEqual(expected.st_ctime, actual.st_ctime)
977
self.assertEqual(expected.st_dev, actual.st_dev)
978
self.assertEqual(expected.st_ino, actual.st_ino)
979
self.assertEqual(expected.st_mode, actual.st_mode)
981
def assertLength(self, length, obj_with_len):
982
"""Assert that obj_with_len is of length length."""
983
if len(obj_with_len) != length:
984
self.fail("Incorrect length: wanted %d, got %d for %r" % (
985
length, len(obj_with_len), obj_with_len))
987
def assertPositive(self, val):
988
"""Assert that val is greater than 0."""
989
self.assertTrue(val > 0, 'expected a positive value, but got %s' % val)
991
def assertNegative(self, val):
992
"""Assert that val is less than 0."""
993
self.assertTrue(val < 0, 'expected a negative value, but got %s' % val)
995
def assertStartsWith(self, s, prefix):
996
if not s.startswith(prefix):
997
raise AssertionError('string %r does not start with %r' % (s, prefix))
999
def assertEndsWith(self, s, suffix):
1000
"""Asserts that s ends with suffix."""
1001
if not s.endswith(suffix):
1002
raise AssertionError('string %r does not end with %r' % (s, suffix))
1004
def assertContainsRe(self, haystack, needle_re, flags=0):
1005
"""Assert that a contains something matching a regular expression."""
1006
if not re.search(needle_re, haystack, flags):
1007
if '\n' in haystack or len(haystack) > 60:
1008
# a long string, format it in a more readable way
1009
raise AssertionError(
1010
'pattern "%s" not found in\n"""\\\n%s"""\n'
1011
% (needle_re, haystack))
1013
raise AssertionError('pattern "%s" not found in "%s"'
1014
% (needle_re, haystack))
1016
def assertNotContainsRe(self, haystack, needle_re, flags=0):
1017
"""Assert that a does not match a regular expression"""
1018
if re.search(needle_re, haystack, flags):
1019
raise AssertionError('pattern "%s" found in "%s"'
1020
% (needle_re, haystack))
1022
def assertSubset(self, sublist, superlist):
1023
"""Assert that every entry in sublist is present in superlist."""
1024
missing = set(sublist) - set(superlist)
1025
if len(missing) > 0:
1026
raise AssertionError("value(s) %r not present in container %r" %
1027
(missing, superlist))
1029
def assertListRaises(self, excClass, func, *args, **kwargs):
1030
"""Fail unless excClass is raised when the iterator from func is used.
1032
Many functions can return generators this makes sure
1033
to wrap them in a list() call to make sure the whole generator
1034
is run, and that the proper exception is raised.
1037
list(func(*args, **kwargs))
1041
if getattr(excClass,'__name__', None) is not None:
1042
excName = excClass.__name__
1044
excName = str(excClass)
1045
raise self.failureException, "%s not raised" % excName
1047
def assertRaises(self, excClass, callableObj, *args, **kwargs):
1048
"""Assert that a callable raises a particular exception.
1050
:param excClass: As for the except statement, this may be either an
1051
exception class, or a tuple of classes.
1052
:param callableObj: A callable, will be passed ``*args`` and
1055
Returns the exception so that you can examine it.
1058
callableObj(*args, **kwargs)
1062
if getattr(excClass,'__name__', None) is not None:
1063
excName = excClass.__name__
1066
excName = str(excClass)
1067
raise self.failureException, "%s not raised" % excName
1069
def assertIs(self, left, right, message=None):
1070
if not (left is right):
1071
if message is not None:
1072
raise AssertionError(message)
1074
raise AssertionError("%r is not %r." % (left, right))
1076
def assertIsNot(self, left, right, message=None):
1078
if message is not None:
1079
raise AssertionError(message)
1081
raise AssertionError("%r is %r." % (left, right))
1083
def assertTransportMode(self, transport, path, mode):
1084
"""Fail if a path does not have mode "mode".
1086
If modes are not supported on this transport, the assertion is ignored.
1088
if not transport._can_roundtrip_unix_modebits():
1090
path_stat = transport.stat(path)
1091
actual_mode = stat.S_IMODE(path_stat.st_mode)
1092
self.assertEqual(mode, actual_mode,
1093
'mode of %r incorrect (%s != %s)'
1094
% (path, oct(mode), oct(actual_mode)))
1096
def assertIsSameRealPath(self, path1, path2):
1097
"""Fail if path1 and path2 points to different files"""
1098
self.assertEqual(osutils.realpath(path1),
1099
osutils.realpath(path2),
1100
"apparent paths:\na = %s\nb = %s\n," % (path1, path2))
1102
def assertIsInstance(self, obj, kls, msg=None):
1103
"""Fail if obj is not an instance of kls
1105
:param msg: Supplementary message to show if the assertion fails.
1107
if not isinstance(obj, kls):
1108
m = "%r is an instance of %s rather than %s" % (
1109
obj, obj.__class__, kls)
1114
def expectFailure(self, reason, assertion, *args, **kwargs):
1115
"""Invoke a test, expecting it to fail for the given reason.
1117
This is for assertions that ought to succeed, but currently fail.
1118
(The failure is *expected* but not *wanted*.) Please be very precise
1119
about the failure you're expecting. If a new bug is introduced,
1120
AssertionError should be raised, not KnownFailure.
1122
Frequently, expectFailure should be followed by an opposite assertion.
1125
Intended to be used with a callable that raises AssertionError as the
1126
'assertion' parameter. args and kwargs are passed to the 'assertion'.
1128
Raises KnownFailure if the test fails. Raises AssertionError if the
1133
self.expectFailure('Math is broken', self.assertNotEqual, 54,
1135
self.assertEqual(42, dynamic_val)
1137
This means that a dynamic_val of 54 will cause the test to raise
1138
a KnownFailure. Once math is fixed and the expectFailure is removed,
1139
only a dynamic_val of 42 will allow the test to pass. Anything other
1140
than 54 or 42 will cause an AssertionError.
1143
assertion(*args, **kwargs)
1144
except AssertionError:
1145
raise KnownFailure(reason)
1147
self.fail('Unexpected success. Should have failed: %s' % reason)
1149
def assertFileEqual(self, content, path):
1150
"""Fail if path does not contain 'content'."""
1151
self.failUnlessExists(path)
1152
f = file(path, 'rb')
1157
self.assertEqualDiff(content, s)
1159
def failUnlessExists(self, path):
1160
"""Fail unless path or paths, which may be abs or relative, exist."""
1161
if not isinstance(path, basestring):
1163
self.failUnlessExists(p)
1165
self.failUnless(osutils.lexists(path),path+" does not exist")
1167
def failIfExists(self, path):
1168
"""Fail if path or paths, which may be abs or relative, exist."""
1169
if not isinstance(path, basestring):
1171
self.failIfExists(p)
1173
self.failIf(osutils.lexists(path),path+" exists")
1175
def _capture_deprecation_warnings(self, a_callable, *args, **kwargs):
1176
"""A helper for callDeprecated and applyDeprecated.
1178
:param a_callable: A callable to call.
1179
:param args: The positional arguments for the callable
1180
:param kwargs: The keyword arguments for the callable
1181
:return: A tuple (warnings, result). result is the result of calling
1182
a_callable(``*args``, ``**kwargs``).
1185
def capture_warnings(msg, cls=None, stacklevel=None):
1186
# we've hooked into a deprecation specific callpath,
1187
# only deprecations should getting sent via it.
1188
self.assertEqual(cls, DeprecationWarning)
1189
local_warnings.append(msg)
1190
original_warning_method = symbol_versioning.warn
1191
symbol_versioning.set_warning_method(capture_warnings)
1193
result = a_callable(*args, **kwargs)
1195
symbol_versioning.set_warning_method(original_warning_method)
1196
return (local_warnings, result)
1198
def applyDeprecated(self, deprecation_format, a_callable, *args, **kwargs):
1199
"""Call a deprecated callable without warning the user.
1201
Note that this only captures warnings raised by symbol_versioning.warn,
1202
not other callers that go direct to the warning module.
1204
To test that a deprecated method raises an error, do something like
1207
self.assertRaises(errors.ReservedId,
1208
self.applyDeprecated,
1209
deprecated_in((1, 5, 0)),
1213
:param deprecation_format: The deprecation format that the callable
1214
should have been deprecated with. This is the same type as the
1215
parameter to deprecated_method/deprecated_function. If the
1216
callable is not deprecated with this format, an assertion error
1218
:param a_callable: A callable to call. This may be a bound method or
1219
a regular function. It will be called with ``*args`` and
1221
:param args: The positional arguments for the callable
1222
:param kwargs: The keyword arguments for the callable
1223
:return: The result of a_callable(``*args``, ``**kwargs``)
1225
call_warnings, result = self._capture_deprecation_warnings(a_callable,
1227
expected_first_warning = symbol_versioning.deprecation_string(
1228
a_callable, deprecation_format)
1229
if len(call_warnings) == 0:
1230
self.fail("No deprecation warning generated by call to %s" %
1232
self.assertEqual(expected_first_warning, call_warnings[0])
1235
def callCatchWarnings(self, fn, *args, **kw):
1236
"""Call a callable that raises python warnings.
1238
The caller's responsible for examining the returned warnings.
1240
If the callable raises an exception, the exception is not
1241
caught and propagates up to the caller. In that case, the list
1242
of warnings is not available.
1244
:returns: ([warning_object, ...], fn_result)
1246
# XXX: This is not perfect, because it completely overrides the
1247
# warnings filters, and some code may depend on suppressing particular
1248
# warnings. It's the easiest way to insulate ourselves from -Werror,
1249
# though. -- Andrew, 20071062
1251
def _catcher(message, category, filename, lineno, file=None, line=None):
1252
# despite the name, 'message' is normally(?) a Warning subclass
1254
wlist.append(message)
1255
saved_showwarning = warnings.showwarning
1256
saved_filters = warnings.filters
1258
warnings.showwarning = _catcher
1259
warnings.filters = []
1260
result = fn(*args, **kw)
1262
warnings.showwarning = saved_showwarning
1263
warnings.filters = saved_filters
1264
return wlist, result
1266
def callDeprecated(self, expected, callable, *args, **kwargs):
1267
"""Assert that a callable is deprecated in a particular way.
1269
This is a very precise test for unusual requirements. The
1270
applyDeprecated helper function is probably more suited for most tests
1271
as it allows you to simply specify the deprecation format being used
1272
and will ensure that that is issued for the function being called.
1274
Note that this only captures warnings raised by symbol_versioning.warn,
1275
not other callers that go direct to the warning module. To catch
1276
general warnings, use callCatchWarnings.
1278
:param expected: a list of the deprecation warnings expected, in order
1279
:param callable: The callable to call
1280
:param args: The positional arguments for the callable
1281
:param kwargs: The keyword arguments for the callable
1283
call_warnings, result = self._capture_deprecation_warnings(callable,
1285
self.assertEqual(expected, call_warnings)
1288
def _startLogFile(self):
1289
"""Send bzr and test log messages to a temporary file.
1291
The file is removed as the test is torn down.
1293
fileno, name = tempfile.mkstemp(suffix='.log', prefix='testbzr')
1294
self._log_file = os.fdopen(fileno, 'w+')
1295
self._log_memento = bzrlib.trace.push_log_file(self._log_file)
1296
self._log_file_name = name
1297
self.addCleanup(self._finishLogFile)
1299
def _finishLogFile(self):
1300
"""Finished with the log file.
1302
Close the file and delete it, unless setKeepLogfile was called.
1304
if self._log_file is None:
1306
bzrlib.trace.pop_log_file(self._log_memento)
1307
self._log_file.close()
1308
self._log_file = None
1309
if not self._keep_log_file:
1310
os.remove(self._log_file_name)
1311
self._log_file_name = None
1313
def setKeepLogfile(self):
1314
"""Make the logfile not be deleted when _finishLogFile is called."""
1315
self._keep_log_file = True
1317
def addCleanup(self, callable, *args, **kwargs):
1318
"""Arrange to run a callable when this case is torn down.
1320
Callables are run in the reverse of the order they are registered,
1321
ie last-in first-out.
1323
self._cleanups.append((callable, args, kwargs))
1325
def _cleanEnvironment(self):
1327
'BZR_HOME': None, # Don't inherit BZR_HOME to all the tests.
1328
'HOME': os.getcwd(),
1329
# bzr now uses the Win32 API and doesn't rely on APPDATA, but the
1330
# tests do check our impls match APPDATA
1331
'BZR_EDITOR': None, # test_msgeditor manipulates this variable
1335
'BZREMAIL': None, # may still be present in the environment
1337
'BZR_PROGRESS_BAR': None,
1339
'BZR_PLUGIN_PATH': None,
1340
# Make sure that any text ui tests are consistent regardless of
1341
# the environment the test case is run in; you may want tests that
1342
# test other combinations. 'dumb' is a reasonable guess for tests
1343
# going to a pipe or a StringIO.
1348
'SSH_AUTH_SOCK': None,
1352
'https_proxy': None,
1353
'HTTPS_PROXY': None,
1358
# Nobody cares about ftp_proxy, FTP_PROXY AFAIK. So far at
1359
# least. If you do (care), please update this comment
1363
'BZR_REMOTE_PATH': None,
1366
self.addCleanup(self._restoreEnvironment)
1367
for name, value in new_env.iteritems():
1368
self._captureVar(name, value)
1370
def _captureVar(self, name, newvalue):
1371
"""Set an environment variable, and reset it when finished."""
1372
self.__old_env[name] = osutils.set_or_unset_env(name, newvalue)
1374
def _restore_debug_flags(self):
1375
debug.debug_flags.clear()
1376
debug.debug_flags.update(self._preserved_debug_flags)
1378
def _restoreEnvironment(self):
1379
for name, value in self.__old_env.iteritems():
1380
osutils.set_or_unset_env(name, value)
1382
def _restoreHooks(self):
1383
for klass, (name, hooks) in self._preserved_hooks.items():
1384
setattr(klass, name, hooks)
1386
def knownFailure(self, reason):
1387
"""This test has failed for some known reason."""
1388
raise KnownFailure(reason)
1390
def _do_skip(self, result, reason):
1391
addSkip = getattr(result, 'addSkip', None)
1392
if not callable(addSkip):
1393
result.addError(self, sys.exc_info())
1395
addSkip(self, reason)
1397
def run(self, result=None):
1398
if result is None: result = self.defaultTestResult()
1399
for feature in getattr(self, '_test_needs_features', []):
1400
if not feature.available():
1401
result.startTest(self)
1402
if getattr(result, 'addNotSupported', None):
1403
result.addNotSupported(self, feature)
1405
result.addSuccess(self)
1406
result.stopTest(self)
1410
result.startTest(self)
1411
absent_attr = object()
1413
method_name = getattr(self, '_testMethodName', absent_attr)
1414
if method_name is absent_attr:
1416
method_name = getattr(self, '_TestCase__testMethodName')
1417
testMethod = getattr(self, method_name)
1421
if not self._bzr_test_setUp_run:
1423
"test setUp did not invoke "
1424
"bzrlib.tests.TestCase's setUp")
1425
except KeyboardInterrupt:
1428
except TestSkipped, e:
1429
self._do_skip(result, e.args[0])
1433
result.addError(self, sys.exc_info())
1441
except self.failureException:
1442
result.addFailure(self, sys.exc_info())
1443
except TestSkipped, e:
1445
reason = "No reason given."
1448
self._do_skip(result, reason)
1449
except KeyboardInterrupt:
1453
result.addError(self, sys.exc_info())
1457
if not self._bzr_test_tearDown_run:
1459
"test tearDown did not invoke "
1460
"bzrlib.tests.TestCase's tearDown")
1461
except KeyboardInterrupt:
1465
result.addError(self, sys.exc_info())
1468
if ok: result.addSuccess(self)
1470
result.stopTest(self)
1472
except TestNotApplicable:
1473
# Not moved from the result [yet].
1476
except KeyboardInterrupt:
1481
for attr_name in self.attrs_to_keep:
1482
if attr_name in self.__dict__:
1483
saved_attrs[attr_name] = self.__dict__[attr_name]
1484
self.__dict__ = saved_attrs
1488
self._log_contents = ''
1489
self._bzr_test_tearDown_run = True
1490
unittest.TestCase.tearDown(self)
1492
def time(self, callable, *args, **kwargs):
1493
"""Run callable and accrue the time it takes to the benchmark time.
1495
If lsprofiling is enabled (i.e. by --lsprof-time to bzr selftest) then
1496
this will cause lsprofile statistics to be gathered and stored in
1499
if self._benchtime is None:
1503
if not self._gather_lsprof_in_benchmarks:
1504
return callable(*args, **kwargs)
1506
# record this benchmark
1507
ret, stats = bzrlib.lsprof.profile(callable, *args, **kwargs)
1509
self._benchcalls.append(((callable, args, kwargs), stats))
1512
self._benchtime += time.time() - start
1514
def _runCleanups(self):
1515
"""Run registered cleanup functions.
1517
This should only be called from TestCase.tearDown.
1519
# TODO: Perhaps this should keep running cleanups even if
1520
# one of them fails?
1522
# Actually pop the cleanups from the list so tearDown running
1523
# twice is safe (this happens for skipped tests).
1524
while self._cleanups:
1525
cleanup, args, kwargs = self._cleanups.pop()
1526
cleanup(*args, **kwargs)
1528
def log(self, *args):
1531
def _get_log(self, keep_log_file=False):
1532
"""Get the log from bzrlib.trace calls from this test.
1534
:param keep_log_file: When True, if the log is still a file on disk
1535
leave it as a file on disk. When False, if the log is still a file
1536
on disk, the log file is deleted and the log preserved as
1538
:return: A string containing the log.
1540
# flush the log file, to get all content
1542
if bzrlib.trace._trace_file:
1543
bzrlib.trace._trace_file.flush()
1544
if self._log_contents:
1545
# XXX: this can hardly contain the content flushed above --vila
1547
return self._log_contents
1548
if self._log_file_name is not None:
1549
logfile = open(self._log_file_name)
1551
log_contents = logfile.read()
1554
if not keep_log_file:
1555
self._log_contents = log_contents
1557
os.remove(self._log_file_name)
1559
if sys.platform == 'win32' and e.errno == errno.EACCES:
1560
sys.stderr.write(('Unable to delete log file '
1561
' %r\n' % self._log_file_name))
1566
return "DELETED log file to reduce memory footprint"
1568
def requireFeature(self, feature):
1569
"""This test requires a specific feature is available.
1571
:raises UnavailableFeature: When feature is not available.
1573
if not feature.available():
1574
raise UnavailableFeature(feature)
1576
def _run_bzr_autosplit(self, args, retcode, encoding, stdin,
1578
"""Run bazaar command line, splitting up a string command line."""
1579
if isinstance(args, basestring):
1580
# shlex don't understand unicode strings,
1581
# so args should be plain string (bialix 20070906)
1582
args = list(shlex.split(str(args)))
1583
return self._run_bzr_core(args, retcode=retcode,
1584
encoding=encoding, stdin=stdin, working_dir=working_dir,
1587
def _run_bzr_core(self, args, retcode, encoding, stdin,
1589
if encoding is None:
1590
encoding = osutils.get_user_encoding()
1591
stdout = StringIOWrapper()
1592
stderr = StringIOWrapper()
1593
stdout.encoding = encoding
1594
stderr.encoding = encoding
1596
self.log('run bzr: %r', args)
1597
# FIXME: don't call into logging here
1598
handler = logging.StreamHandler(stderr)
1599
handler.setLevel(logging.INFO)
1600
logger = logging.getLogger('')
1601
logger.addHandler(handler)
1602
old_ui_factory = ui.ui_factory
1603
ui.ui_factory = TestUIFactory(stdin=stdin, stdout=stdout, stderr=stderr)
1606
if working_dir is not None:
1607
cwd = osutils.getcwd()
1608
os.chdir(working_dir)
1611
result = self.apply_redirected(ui.ui_factory.stdin,
1613
bzrlib.commands.run_bzr_catch_user_errors,
1616
logger.removeHandler(handler)
1617
ui.ui_factory = old_ui_factory
1621
out = stdout.getvalue()
1622
err = stderr.getvalue()
1624
self.log('output:\n%r', out)
1626
self.log('errors:\n%r', err)
1627
if retcode is not None:
1628
self.assertEquals(retcode, result,
1629
message='Unexpected return code')
1632
def run_bzr(self, args, retcode=0, encoding=None, stdin=None,
1633
working_dir=None, error_regexes=[], output_encoding=None):
1634
"""Invoke bzr, as if it were run from the command line.
1636
The argument list should not include the bzr program name - the
1637
first argument is normally the bzr command. Arguments may be
1638
passed in three ways:
1640
1- A list of strings, eg ["commit", "a"]. This is recommended
1641
when the command contains whitespace or metacharacters, or
1642
is built up at run time.
1644
2- A single string, eg "add a". This is the most convenient
1645
for hardcoded commands.
1647
This runs bzr through the interface that catches and reports
1648
errors, and with logging set to something approximating the
1649
default, so that error reporting can be checked.
1651
This should be the main method for tests that want to exercise the
1652
overall behavior of the bzr application (rather than a unit test
1653
or a functional test of the library.)
1655
This sends the stdout/stderr results into the test's log,
1656
where it may be useful for debugging. See also run_captured.
1658
:keyword stdin: A string to be used as stdin for the command.
1659
:keyword retcode: The status code the command should return;
1661
:keyword working_dir: The directory to run the command in
1662
:keyword error_regexes: A list of expected error messages. If
1663
specified they must be seen in the error output of the command.
1665
out, err = self._run_bzr_autosplit(
1670
working_dir=working_dir,
1672
self.assertIsInstance(error_regexes, (list, tuple))
1673
for regex in error_regexes:
1674
self.assertContainsRe(err, regex)
1677
def run_bzr_error(self, error_regexes, *args, **kwargs):
1678
"""Run bzr, and check that stderr contains the supplied regexes
1680
:param error_regexes: Sequence of regular expressions which
1681
must each be found in the error output. The relative ordering
1683
:param args: command-line arguments for bzr
1684
:param kwargs: Keyword arguments which are interpreted by run_bzr
1685
This function changes the default value of retcode to be 3,
1686
since in most cases this is run when you expect bzr to fail.
1688
:return: (out, err) The actual output of running the command (in case
1689
you want to do more inspection)
1693
# Make sure that commit is failing because there is nothing to do
1694
self.run_bzr_error(['no changes to commit'],
1695
['commit', '-m', 'my commit comment'])
1696
# Make sure --strict is handling an unknown file, rather than
1697
# giving us the 'nothing to do' error
1698
self.build_tree(['unknown'])
1699
self.run_bzr_error(['Commit refused because there are unknown files'],
1700
['commit', --strict', '-m', 'my commit comment'])
1702
kwargs.setdefault('retcode', 3)
1703
kwargs['error_regexes'] = error_regexes
1704
out, err = self.run_bzr(*args, **kwargs)
1707
def run_bzr_subprocess(self, *args, **kwargs):
1708
"""Run bzr in a subprocess for testing.
1710
This starts a new Python interpreter and runs bzr in there.
1711
This should only be used for tests that have a justifiable need for
1712
this isolation: e.g. they are testing startup time, or signal
1713
handling, or early startup code, etc. Subprocess code can't be
1714
profiled or debugged so easily.
1716
:keyword retcode: The status code that is expected. Defaults to 0. If
1717
None is supplied, the status code is not checked.
1718
:keyword env_changes: A dictionary which lists changes to environment
1719
variables. A value of None will unset the env variable.
1720
The values must be strings. The change will only occur in the
1721
child, so you don't need to fix the environment after running.
1722
:keyword universal_newlines: Convert CRLF => LF
1723
:keyword allow_plugins: By default the subprocess is run with
1724
--no-plugins to ensure test reproducibility. Also, it is possible
1725
for system-wide plugins to create unexpected output on stderr,
1726
which can cause unnecessary test failures.
1728
env_changes = kwargs.get('env_changes', {})
1729
working_dir = kwargs.get('working_dir', None)
1730
allow_plugins = kwargs.get('allow_plugins', False)
1732
if isinstance(args[0], list):
1734
elif isinstance(args[0], basestring):
1735
args = list(shlex.split(args[0]))
1737
raise ValueError("passing varargs to run_bzr_subprocess")
1738
process = self.start_bzr_subprocess(args, env_changes=env_changes,
1739
working_dir=working_dir,
1740
allow_plugins=allow_plugins)
1741
# We distinguish between retcode=None and retcode not passed.
1742
supplied_retcode = kwargs.get('retcode', 0)
1743
return self.finish_bzr_subprocess(process, retcode=supplied_retcode,
1744
universal_newlines=kwargs.get('universal_newlines', False),
1747
def start_bzr_subprocess(self, process_args, env_changes=None,
1748
skip_if_plan_to_signal=False,
1750
allow_plugins=False):
1751
"""Start bzr in a subprocess for testing.
1753
This starts a new Python interpreter and runs bzr in there.
1754
This should only be used for tests that have a justifiable need for
1755
this isolation: e.g. they are testing startup time, or signal
1756
handling, or early startup code, etc. Subprocess code can't be
1757
profiled or debugged so easily.
1759
:param process_args: a list of arguments to pass to the bzr executable,
1760
for example ``['--version']``.
1761
:param env_changes: A dictionary which lists changes to environment
1762
variables. A value of None will unset the env variable.
1763
The values must be strings. The change will only occur in the
1764
child, so you don't need to fix the environment after running.
1765
:param skip_if_plan_to_signal: raise TestSkipped when true and os.kill
1767
:param allow_plugins: If False (default) pass --no-plugins to bzr.
1769
:returns: Popen object for the started process.
1771
if skip_if_plan_to_signal:
1772
if not getattr(os, 'kill', None):
1773
raise TestSkipped("os.kill not available.")
1775
if env_changes is None:
1779
def cleanup_environment():
1780
for env_var, value in env_changes.iteritems():
1781
old_env[env_var] = osutils.set_or_unset_env(env_var, value)
1783
def restore_environment():
1784
for env_var, value in old_env.iteritems():
1785
osutils.set_or_unset_env(env_var, value)
1787
bzr_path = self.get_bzr_path()
1790
if working_dir is not None:
1791
cwd = osutils.getcwd()
1792
os.chdir(working_dir)
1795
# win32 subprocess doesn't support preexec_fn
1796
# so we will avoid using it on all platforms, just to
1797
# make sure the code path is used, and we don't break on win32
1798
cleanup_environment()
1799
command = [sys.executable]
1800
# frozen executables don't need the path to bzr
1801
if getattr(sys, "frozen", None) is None:
1802
command.append(bzr_path)
1803
if not allow_plugins:
1804
command.append('--no-plugins')
1805
command.extend(process_args)
1806
process = self._popen(command, stdin=PIPE, stdout=PIPE, stderr=PIPE)
1808
restore_environment()
1814
def _popen(self, *args, **kwargs):
1815
"""Place a call to Popen.
1817
Allows tests to override this method to intercept the calls made to
1818
Popen for introspection.
1820
return Popen(*args, **kwargs)
1822
def get_bzr_path(self):
1823
"""Return the path of the 'bzr' executable for this test suite."""
1824
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
1825
if not os.path.isfile(bzr_path):
1826
# We are probably installed. Assume sys.argv is the right file
1827
bzr_path = sys.argv[0]
1830
def finish_bzr_subprocess(self, process, retcode=0, send_signal=None,
1831
universal_newlines=False, process_args=None):
1832
"""Finish the execution of process.
1834
:param process: the Popen object returned from start_bzr_subprocess.
1835
:param retcode: The status code that is expected. Defaults to 0. If
1836
None is supplied, the status code is not checked.
1837
:param send_signal: an optional signal to send to the process.
1838
:param universal_newlines: Convert CRLF => LF
1839
:returns: (stdout, stderr)
1841
if send_signal is not None:
1842
os.kill(process.pid, send_signal)
1843
out, err = process.communicate()
1845
if universal_newlines:
1846
out = out.replace('\r\n', '\n')
1847
err = err.replace('\r\n', '\n')
1849
if retcode is not None and retcode != process.returncode:
1850
if process_args is None:
1851
process_args = "(unknown args)"
1852
mutter('Output of bzr %s:\n%s', process_args, out)
1853
mutter('Error for bzr %s:\n%s', process_args, err)
1854
self.fail('Command bzr %s failed with retcode %s != %s'
1855
% (process_args, retcode, process.returncode))
1858
def check_inventory_shape(self, inv, shape):
1859
"""Compare an inventory to a list of expected names.
1861
Fail if they are not precisely equal.
1864
shape = list(shape) # copy
1865
for path, ie in inv.entries():
1866
name = path.replace('\\', '/')
1867
if ie.kind == 'directory':
1874
self.fail("expected paths not found in inventory: %r" % shape)
1876
self.fail("unexpected paths found in inventory: %r" % extras)
1878
def apply_redirected(self, stdin=None, stdout=None, stderr=None,
1879
a_callable=None, *args, **kwargs):
1880
"""Call callable with redirected std io pipes.
1882
Returns the return code."""
1883
if not callable(a_callable):
1884
raise ValueError("a_callable must be callable.")
1886
stdin = StringIO("")
1888
if getattr(self, "_log_file", None) is not None:
1889
stdout = self._log_file
1893
if getattr(self, "_log_file", None is not None):
1894
stderr = self._log_file
1897
real_stdin = sys.stdin
1898
real_stdout = sys.stdout
1899
real_stderr = sys.stderr
1904
return a_callable(*args, **kwargs)
1906
sys.stdout = real_stdout
1907
sys.stderr = real_stderr
1908
sys.stdin = real_stdin
1910
def reduceLockdirTimeout(self):
1911
"""Reduce the default lock timeout for the duration of the test, so that
1912
if LockContention occurs during a test, it does so quickly.
1914
Tests that expect to provoke LockContention errors should call this.
1916
orig_timeout = bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS
1918
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = orig_timeout
1919
self.addCleanup(resetTimeout)
1920
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = 0
1922
def make_utf8_encoded_stringio(self, encoding_type=None):
1923
"""Return a StringIOWrapper instance, that will encode Unicode
1926
if encoding_type is None:
1927
encoding_type = 'strict'
1929
output_encoding = 'utf-8'
1930
sio = codecs.getwriter(output_encoding)(sio, errors=encoding_type)
1931
sio.encoding = output_encoding
1935
class CapturedCall(object):
1936
"""A helper for capturing smart server calls for easy debug analysis."""
1938
def __init__(self, params, prefix_length):
1939
"""Capture the call with params and skip prefix_length stack frames."""
1942
# The last 5 frames are the __init__, the hook frame, and 3 smart
1943
# client frames. Beyond this we could get more clever, but this is good
1945
stack = traceback.extract_stack()[prefix_length:-5]
1946
self.stack = ''.join(traceback.format_list(stack))
1949
return self.call.method
1952
return self.call.method
1958
class TestCaseWithMemoryTransport(TestCase):
1959
"""Common test class for tests that do not need disk resources.
1961
Tests that need disk resources should derive from TestCaseWithTransport.
1963
TestCaseWithMemoryTransport sets the TEST_ROOT variable for all bzr tests.
1965
For TestCaseWithMemoryTransport the test_home_dir is set to the name of
1966
a directory which does not exist. This serves to help ensure test isolation
1967
is preserved. test_dir is set to the TEST_ROOT, as is cwd, because they
1968
must exist. However, TestCaseWithMemoryTransport does not offer local
1969
file defaults for the transport in tests, nor does it obey the command line
1970
override, so tests that accidentally write to the common directory should
1973
:cvar TEST_ROOT: Directory containing all temporary directories, plus
1974
a .bzr directory that stops us ascending higher into the filesystem.
1980
def __init__(self, methodName='runTest'):
1981
# allow test parameterization after test construction and before test
1982
# execution. Variables that the parameterizer sets need to be
1983
# ones that are not set by setUp, or setUp will trash them.
1984
super(TestCaseWithMemoryTransport, self).__init__(methodName)
1985
self.vfs_transport_factory = default_transport
1986
self.transport_server = None
1987
self.transport_readonly_server = None
1988
self.__vfs_server = None
1990
def get_transport(self, relpath=None):
1991
"""Return a writeable transport.
1993
This transport is for the test scratch space relative to
1996
:param relpath: a path relative to the base url.
1998
t = get_transport(self.get_url(relpath))
1999
self.assertFalse(t.is_readonly())
2002
def get_readonly_transport(self, relpath=None):
2003
"""Return a readonly transport for the test scratch space
2005
This can be used to test that operations which should only need
2006
readonly access in fact do not try to write.
2008
:param relpath: a path relative to the base url.
2010
t = get_transport(self.get_readonly_url(relpath))
2011
self.assertTrue(t.is_readonly())
2014
def create_transport_readonly_server(self):
2015
"""Create a transport server from class defined at init.
2017
This is mostly a hook for daughter classes.
2019
return self.transport_readonly_server()
2021
def get_readonly_server(self):
2022
"""Get the server instance for the readonly transport
2024
This is useful for some tests with specific servers to do diagnostics.
2026
if self.__readonly_server is None:
2027
if self.transport_readonly_server is None:
2028
# readonly decorator requested
2029
# bring up the server
2030
self.__readonly_server = ReadonlyServer()
2031
self.__readonly_server.setUp(self.get_vfs_only_server())
2033
self.__readonly_server = self.create_transport_readonly_server()
2034
self.__readonly_server.setUp(self.get_vfs_only_server())
2035
self.addCleanup(self.__readonly_server.tearDown)
2036
return self.__readonly_server
2038
def get_readonly_url(self, relpath=None):
2039
"""Get a URL for the readonly transport.
2041
This will either be backed by '.' or a decorator to the transport
2042
used by self.get_url()
2043
relpath provides for clients to get a path relative to the base url.
2044
These should only be downwards relative, not upwards.
2046
base = self.get_readonly_server().get_url()
2047
return self._adjust_url(base, relpath)
2049
def get_vfs_only_server(self):
2050
"""Get the vfs only read/write server instance.
2052
This is useful for some tests with specific servers that need
2055
For TestCaseWithMemoryTransport this is always a MemoryServer, and there
2056
is no means to override it.
2058
if self.__vfs_server is None:
2059
self.__vfs_server = MemoryServer()
2060
self.__vfs_server.setUp()
2061
self.addCleanup(self.__vfs_server.tearDown)
2062
return self.__vfs_server
2064
def get_server(self):
2065
"""Get the read/write server instance.
2067
This is useful for some tests with specific servers that need
2070
This is built from the self.transport_server factory. If that is None,
2071
then the self.get_vfs_server is returned.
2073
if self.__server is None:
2074
if self.transport_server is None or self.transport_server is self.vfs_transport_factory:
2075
return self.get_vfs_only_server()
2077
# bring up a decorated means of access to the vfs only server.
2078
self.__server = self.transport_server()
2080
self.__server.setUp(self.get_vfs_only_server())
2081
except TypeError, e:
2082
# This should never happen; the try:Except here is to assist
2083
# developers having to update code rather than seeing an
2084
# uninformative TypeError.
2085
raise Exception, "Old server API in use: %s, %s" % (self.__server, e)
2086
self.addCleanup(self.__server.tearDown)
2087
return self.__server
2089
def _adjust_url(self, base, relpath):
2090
"""Get a URL (or maybe a path) for the readwrite transport.
2092
This will either be backed by '.' or to an equivalent non-file based
2094
relpath provides for clients to get a path relative to the base url.
2095
These should only be downwards relative, not upwards.
2097
if relpath is not None and relpath != '.':
2098
if not base.endswith('/'):
2100
# XXX: Really base should be a url; we did after all call
2101
# get_url()! But sometimes it's just a path (from
2102
# LocalAbspathServer), and it'd be wrong to append urlescaped data
2103
# to a non-escaped local path.
2104
if base.startswith('./') or base.startswith('/'):
2107
base += urlutils.escape(relpath)
2110
def get_url(self, relpath=None):
2111
"""Get a URL (or maybe a path) for the readwrite transport.
2113
This will either be backed by '.' or to an equivalent non-file based
2115
relpath provides for clients to get a path relative to the base url.
2116
These should only be downwards relative, not upwards.
2118
base = self.get_server().get_url()
2119
return self._adjust_url(base, relpath)
2121
def get_vfs_only_url(self, relpath=None):
2122
"""Get a URL (or maybe a path for the plain old vfs transport.
2124
This will never be a smart protocol. It always has all the
2125
capabilities of the local filesystem, but it might actually be a
2126
MemoryTransport or some other similar virtual filesystem.
2128
This is the backing transport (if any) of the server returned by
2129
get_url and get_readonly_url.
2131
:param relpath: provides for clients to get a path relative to the base
2132
url. These should only be downwards relative, not upwards.
2135
base = self.get_vfs_only_server().get_url()
2136
return self._adjust_url(base, relpath)
2138
def _create_safety_net(self):
2139
"""Make a fake bzr directory.
2141
This prevents any tests propagating up onto the TEST_ROOT directory's
2144
root = TestCaseWithMemoryTransport.TEST_ROOT
2145
bzrdir.BzrDir.create_standalone_workingtree(root)
2147
def _check_safety_net(self):
2148
"""Check that the safety .bzr directory have not been touched.
2150
_make_test_root have created a .bzr directory to prevent tests from
2151
propagating. This method ensures than a test did not leaked.
2153
root = TestCaseWithMemoryTransport.TEST_ROOT
2154
wt = workingtree.WorkingTree.open(root)
2155
last_rev = wt.last_revision()
2156
if last_rev != 'null:':
2157
# The current test have modified the /bzr directory, we need to
2158
# recreate a new one or all the followng tests will fail.
2159
# If you need to inspect its content uncomment the following line
2160
# import pdb; pdb.set_trace()
2161
_rmtree_temp_dir(root + '/.bzr')
2162
self._create_safety_net()
2163
raise AssertionError('%s/.bzr should not be modified' % root)
2165
def _make_test_root(self):
2166
if TestCaseWithMemoryTransport.TEST_ROOT is None:
2167
root = osutils.mkdtemp(prefix='testbzr-', suffix='.tmp')
2168
TestCaseWithMemoryTransport.TEST_ROOT = root
2170
self._create_safety_net()
2172
# The same directory is used by all tests, and we're not
2173
# specifically told when all tests are finished. This will do.
2174
atexit.register(_rmtree_temp_dir, root)
2176
self.addCleanup(self._check_safety_net)
2178
def makeAndChdirToTestDir(self):
2179
"""Create a temporary directories for this one test.
2181
This must set self.test_home_dir and self.test_dir and chdir to
2184
For TestCaseWithMemoryTransport we chdir to the TEST_ROOT for this test.
2186
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2187
self.test_dir = TestCaseWithMemoryTransport.TEST_ROOT
2188
self.test_home_dir = self.test_dir + "/MemoryTransportMissingHomeDir"
2190
def make_branch(self, relpath, format=None):
2191
"""Create a branch on the transport at relpath."""
2192
repo = self.make_repository(relpath, format=format)
2193
return repo.bzrdir.create_branch()
2195
def make_bzrdir(self, relpath, format=None):
2197
# might be a relative or absolute path
2198
maybe_a_url = self.get_url(relpath)
2199
segments = maybe_a_url.rsplit('/', 1)
2200
t = get_transport(maybe_a_url)
2201
if len(segments) > 1 and segments[-1] not in ('', '.'):
2205
if isinstance(format, basestring):
2206
format = bzrdir.format_registry.make_bzrdir(format)
2207
return format.initialize_on_transport(t)
2208
except errors.UninitializableFormat:
2209
raise TestSkipped("Format %s is not initializable." % format)
2211
def make_repository(self, relpath, shared=False, format=None):
2212
"""Create a repository on our default transport at relpath.
2214
Note that relpath must be a relative path, not a full url.
2216
# FIXME: If you create a remoterepository this returns the underlying
2217
# real format, which is incorrect. Actually we should make sure that
2218
# RemoteBzrDir returns a RemoteRepository.
2219
# maybe mbp 20070410
2220
made_control = self.make_bzrdir(relpath, format=format)
2221
return made_control.create_repository(shared=shared)
2223
def make_smart_server(self, path):
2224
smart_server = server.SmartTCPServer_for_testing()
2225
smart_server.setUp(self.get_server())
2226
remote_transport = get_transport(smart_server.get_url()).clone(path)
2227
self.addCleanup(smart_server.tearDown)
2228
return remote_transport
2230
def make_branch_and_memory_tree(self, relpath, format=None):
2231
"""Create a branch on the default transport and a MemoryTree for it."""
2232
b = self.make_branch(relpath, format=format)
2233
return memorytree.MemoryTree.create_on_branch(b)
2235
def make_branch_builder(self, relpath, format=None):
2236
branch = self.make_branch(relpath, format=format)
2237
return branchbuilder.BranchBuilder(branch=branch)
2239
def overrideEnvironmentForTesting(self):
2240
os.environ['HOME'] = self.test_home_dir
2241
os.environ['BZR_HOME'] = self.test_home_dir
2244
super(TestCaseWithMemoryTransport, self).setUp()
2245
self._make_test_root()
2246
_currentdir = os.getcwdu()
2247
def _leaveDirectory():
2248
os.chdir(_currentdir)
2249
self.addCleanup(_leaveDirectory)
2250
self.makeAndChdirToTestDir()
2251
self.overrideEnvironmentForTesting()
2252
self.__readonly_server = None
2253
self.__server = None
2254
self.reduceLockdirTimeout()
2256
def setup_smart_server_with_call_log(self):
2257
"""Sets up a smart server as the transport server with a call log."""
2258
self.transport_server = server.SmartTCPServer_for_testing
2259
self.hpss_calls = []
2261
# Skip the current stack down to the caller of
2262
# setup_smart_server_with_call_log
2263
prefix_length = len(traceback.extract_stack()) - 2
2264
def capture_hpss_call(params):
2265
self.hpss_calls.append(
2266
CapturedCall(params, prefix_length))
2267
client._SmartClient.hooks.install_named_hook(
2268
'call', capture_hpss_call, None)
2270
def reset_smart_call_log(self):
2271
self.hpss_calls = []
2274
class TestCaseInTempDir(TestCaseWithMemoryTransport):
2275
"""Derived class that runs a test within a temporary directory.
2277
This is useful for tests that need to create a branch, etc.
2279
The directory is created in a slightly complex way: for each
2280
Python invocation, a new temporary top-level directory is created.
2281
All test cases create their own directory within that. If the
2282
tests complete successfully, the directory is removed.
2284
:ivar test_base_dir: The path of the top-level directory for this
2285
test, which contains a home directory and a work directory.
2287
:ivar test_home_dir: An initially empty directory under test_base_dir
2288
which is used as $HOME for this test.
2290
:ivar test_dir: A directory under test_base_dir used as the current
2291
directory when the test proper is run.
2294
OVERRIDE_PYTHON = 'python'
2296
def check_file_contents(self, filename, expect):
2297
self.log("check contents of file %s" % filename)
2298
contents = file(filename, 'r').read()
2299
if contents != expect:
2300
self.log("expected: %r" % expect)
2301
self.log("actually: %r" % contents)
2302
self.fail("contents of %s not as expected" % filename)
2304
def _getTestDirPrefix(self):
2305
# create a directory within the top level test directory
2306
if sys.platform == 'win32':
2307
name_prefix = re.sub('[<>*=+",:;_/\\-]', '_', self.id())
2308
# windows is likely to have path-length limits so use a short name
2309
name_prefix = name_prefix[-30:]
2311
name_prefix = re.sub('[/]', '_', self.id())
2314
def makeAndChdirToTestDir(self):
2315
"""See TestCaseWithMemoryTransport.makeAndChdirToTestDir().
2317
For TestCaseInTempDir we create a temporary directory based on the test
2318
name and then create two subdirs - test and home under it.
2320
name_prefix = osutils.pathjoin(TestCaseWithMemoryTransport.TEST_ROOT,
2321
self._getTestDirPrefix())
2323
for i in range(100):
2324
if os.path.exists(name):
2325
name = name_prefix + '_' + str(i)
2329
# now create test and home directories within this dir
2330
self.test_base_dir = name
2331
self.test_home_dir = self.test_base_dir + '/home'
2332
os.mkdir(self.test_home_dir)
2333
self.test_dir = self.test_base_dir + '/work'
2334
os.mkdir(self.test_dir)
2335
os.chdir(self.test_dir)
2336
# put name of test inside
2337
f = file(self.test_base_dir + '/name', 'w')
2342
self.addCleanup(self.deleteTestDir)
2344
def deleteTestDir(self):
2345
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2346
_rmtree_temp_dir(self.test_base_dir)
2348
def build_tree(self, shape, line_endings='binary', transport=None):
2349
"""Build a test tree according to a pattern.
2351
shape is a sequence of file specifications. If the final
2352
character is '/', a directory is created.
2354
This assumes that all the elements in the tree being built are new.
2356
This doesn't add anything to a branch.
2358
:type shape: list or tuple.
2359
:param line_endings: Either 'binary' or 'native'
2360
in binary mode, exact contents are written in native mode, the
2361
line endings match the default platform endings.
2362
:param transport: A transport to write to, for building trees on VFS's.
2363
If the transport is readonly or None, "." is opened automatically.
2366
if type(shape) not in (list, tuple):
2367
raise AssertionError("Parameter 'shape' should be "
2368
"a list or a tuple. Got %r instead" % (shape,))
2369
# It's OK to just create them using forward slashes on windows.
2370
if transport is None or transport.is_readonly():
2371
transport = get_transport(".")
2373
self.assertIsInstance(name, basestring)
2375
transport.mkdir(urlutils.escape(name[:-1]))
2377
if line_endings == 'binary':
2379
elif line_endings == 'native':
2382
raise errors.BzrError(
2383
'Invalid line ending request %r' % line_endings)
2384
content = "contents of %s%s" % (name.encode('utf-8'), end)
2385
transport.put_bytes_non_atomic(urlutils.escape(name), content)
2387
def build_tree_contents(self, shape):
2388
build_tree_contents(shape)
2390
def assertInWorkingTree(self, path, root_path='.', tree=None):
2391
"""Assert whether path or paths are in the WorkingTree"""
2393
tree = workingtree.WorkingTree.open(root_path)
2394
if not isinstance(path, basestring):
2396
self.assertInWorkingTree(p, tree=tree)
2398
self.assertIsNot(tree.path2id(path), None,
2399
path+' not in working tree.')
2401
def assertNotInWorkingTree(self, path, root_path='.', tree=None):
2402
"""Assert whether path or paths are not in the WorkingTree"""
2404
tree = workingtree.WorkingTree.open(root_path)
2405
if not isinstance(path, basestring):
2407
self.assertNotInWorkingTree(p,tree=tree)
2409
self.assertIs(tree.path2id(path), None, path+' in working tree.')
2412
class TestCaseWithTransport(TestCaseInTempDir):
2413
"""A test case that provides get_url and get_readonly_url facilities.
2415
These back onto two transport servers, one for readonly access and one for
2418
If no explicit class is provided for readonly access, a
2419
ReadonlyTransportDecorator is used instead which allows the use of non disk
2420
based read write transports.
2422
If an explicit class is provided for readonly access, that server and the
2423
readwrite one must both define get_url() as resolving to os.getcwd().
2426
def get_vfs_only_server(self):
2427
"""See TestCaseWithMemoryTransport.
2429
This is useful for some tests with specific servers that need
2432
if self.__vfs_server is None:
2433
self.__vfs_server = self.vfs_transport_factory()
2434
self.__vfs_server.setUp()
2435
self.addCleanup(self.__vfs_server.tearDown)
2436
return self.__vfs_server
2438
def make_branch_and_tree(self, relpath, format=None):
2439
"""Create a branch on the transport and a tree locally.
2441
If the transport is not a LocalTransport, the Tree can't be created on
2442
the transport. In that case if the vfs_transport_factory is
2443
LocalURLServer the working tree is created in the local
2444
directory backing the transport, and the returned tree's branch and
2445
repository will also be accessed locally. Otherwise a lightweight
2446
checkout is created and returned.
2448
:param format: The BzrDirFormat.
2449
:returns: the WorkingTree.
2451
# TODO: always use the local disk path for the working tree,
2452
# this obviously requires a format that supports branch references
2453
# so check for that by checking bzrdir.BzrDirFormat.get_default_format()
2455
b = self.make_branch(relpath, format=format)
2457
return b.bzrdir.create_workingtree()
2458
except errors.NotLocalUrl:
2459
# We can only make working trees locally at the moment. If the
2460
# transport can't support them, then we keep the non-disk-backed
2461
# branch and create a local checkout.
2462
if self.vfs_transport_factory is LocalURLServer:
2463
# the branch is colocated on disk, we cannot create a checkout.
2464
# hopefully callers will expect this.
2465
local_controldir= bzrdir.BzrDir.open(self.get_vfs_only_url(relpath))
2466
wt = local_controldir.create_workingtree()
2467
if wt.branch._format != b._format:
2469
# Make sure that assigning to wt._branch fixes wt.branch,
2470
# in case the implementation details of workingtree objects
2472
self.assertIs(b, wt.branch)
2475
return b.create_checkout(relpath, lightweight=True)
2477
def assertIsDirectory(self, relpath, transport):
2478
"""Assert that relpath within transport is a directory.
2480
This may not be possible on all transports; in that case it propagates
2481
a TransportNotPossible.
2484
mode = transport.stat(relpath).st_mode
2485
except errors.NoSuchFile:
2486
self.fail("path %s is not a directory; no such file"
2488
if not stat.S_ISDIR(mode):
2489
self.fail("path %s is not a directory; has mode %#o"
2492
def assertTreesEqual(self, left, right):
2493
"""Check that left and right have the same content and properties."""
2494
# we use a tree delta to check for equality of the content, and we
2495
# manually check for equality of other things such as the parents list.
2496
self.assertEqual(left.get_parent_ids(), right.get_parent_ids())
2497
differences = left.changes_from(right)
2498
self.assertFalse(differences.has_changed(),
2499
"Trees %r and %r are different: %r" % (left, right, differences))
2502
super(TestCaseWithTransport, self).setUp()
2503
self.__vfs_server = None
2506
class ChrootedTestCase(TestCaseWithTransport):
2507
"""A support class that provides readonly urls outside the local namespace.
2509
This is done by checking if self.transport_server is a MemoryServer. if it
2510
is then we are chrooted already, if it is not then an HttpServer is used
2513
TODO RBC 20060127: make this an option to TestCaseWithTransport so it can
2514
be used without needed to redo it when a different
2515
subclass is in use ?
2519
super(ChrootedTestCase, self).setUp()
2520
if not self.vfs_transport_factory == MemoryServer:
2521
self.transport_readonly_server = HttpServer
2524
def condition_id_re(pattern):
2525
"""Create a condition filter which performs a re check on a test's id.
2527
:param pattern: A regular expression string.
2528
:return: A callable that returns True if the re matches.
2530
filter_re = osutils.re_compile_checked(pattern, 0,
2532
def condition(test):
2534
return filter_re.search(test_id)
2538
def condition_isinstance(klass_or_klass_list):
2539
"""Create a condition filter which returns isinstance(param, klass).
2541
:return: A callable which when called with one parameter obj return the
2542
result of isinstance(obj, klass_or_klass_list).
2545
return isinstance(obj, klass_or_klass_list)
2549
def condition_id_in_list(id_list):
2550
"""Create a condition filter which verify that test's id in a list.
2552
:param id_list: A TestIdList object.
2553
:return: A callable that returns True if the test's id appears in the list.
2555
def condition(test):
2556
return id_list.includes(test.id())
2560
def condition_id_startswith(starts):
2561
"""Create a condition filter verifying that test's id starts with a string.
2563
:param starts: A list of string.
2564
:return: A callable that returns True if the test's id starts with one of
2567
def condition(test):
2568
for start in starts:
2569
if test.id().startswith(start):
2575
def exclude_tests_by_condition(suite, condition):
2576
"""Create a test suite which excludes some tests from suite.
2578
:param suite: The suite to get tests from.
2579
:param condition: A callable whose result evaluates True when called with a
2580
test case which should be excluded from the result.
2581
:return: A suite which contains the tests found in suite that fail
2585
for test in iter_suite_tests(suite):
2586
if not condition(test):
2588
return TestUtil.TestSuite(result)
2591
def filter_suite_by_condition(suite, condition):
2592
"""Create a test suite by filtering another one.
2594
:param suite: The source suite.
2595
:param condition: A callable whose result evaluates True when called with a
2596
test case which should be included in the result.
2597
:return: A suite which contains the tests found in suite that pass
2601
for test in iter_suite_tests(suite):
2604
return TestUtil.TestSuite(result)
2607
def filter_suite_by_re(suite, pattern):
2608
"""Create a test suite by filtering another one.
2610
:param suite: the source suite
2611
:param pattern: pattern that names must match
2612
:returns: the newly created suite
2614
condition = condition_id_re(pattern)
2615
result_suite = filter_suite_by_condition(suite, condition)
2619
def filter_suite_by_id_list(suite, test_id_list):
2620
"""Create a test suite by filtering another one.
2622
:param suite: The source suite.
2623
:param test_id_list: A list of the test ids to keep as strings.
2624
:returns: the newly created suite
2626
condition = condition_id_in_list(test_id_list)
2627
result_suite = filter_suite_by_condition(suite, condition)
2631
def filter_suite_by_id_startswith(suite, start):
2632
"""Create a test suite by filtering another one.
2634
:param suite: The source suite.
2635
:param start: A list of string the test id must start with one of.
2636
:returns: the newly created suite
2638
condition = condition_id_startswith(start)
2639
result_suite = filter_suite_by_condition(suite, condition)
2643
def exclude_tests_by_re(suite, pattern):
2644
"""Create a test suite which excludes some tests from suite.
2646
:param suite: The suite to get tests from.
2647
:param pattern: A regular expression string. Test ids that match this
2648
pattern will be excluded from the result.
2649
:return: A TestSuite that contains all the tests from suite without the
2650
tests that matched pattern. The order of tests is the same as it was in
2653
return exclude_tests_by_condition(suite, condition_id_re(pattern))
2656
def preserve_input(something):
2657
"""A helper for performing test suite transformation chains.
2659
:param something: Anything you want to preserve.
2665
def randomize_suite(suite):
2666
"""Return a new TestSuite with suite's tests in random order.
2668
The tests in the input suite are flattened into a single suite in order to
2669
accomplish this. Any nested TestSuites are removed to provide global
2672
tests = list(iter_suite_tests(suite))
2673
random.shuffle(tests)
2674
return TestUtil.TestSuite(tests)
2677
def split_suite_by_condition(suite, condition):
2678
"""Split a test suite into two by a condition.
2680
:param suite: The suite to split.
2681
:param condition: The condition to match on. Tests that match this
2682
condition are returned in the first test suite, ones that do not match
2683
are in the second suite.
2684
:return: A tuple of two test suites, where the first contains tests from
2685
suite matching the condition, and the second contains the remainder
2686
from suite. The order within each output suite is the same as it was in
2691
for test in iter_suite_tests(suite):
2693
matched.append(test)
2695
did_not_match.append(test)
2696
return TestUtil.TestSuite(matched), TestUtil.TestSuite(did_not_match)
2699
def split_suite_by_re(suite, pattern):
2700
"""Split a test suite into two by a regular expression.
2702
:param suite: The suite to split.
2703
:param pattern: A regular expression string. Test ids that match this
2704
pattern will be in the first test suite returned, and the others in the
2705
second test suite returned.
2706
:return: A tuple of two test suites, where the first contains tests from
2707
suite matching pattern, and the second contains the remainder from
2708
suite. The order within each output suite is the same as it was in
2711
return split_suite_by_condition(suite, condition_id_re(pattern))
2714
def run_suite(suite, name='test', verbose=False, pattern=".*",
2715
stop_on_failure=False,
2716
transport=None, lsprof_timed=None, bench_history=None,
2717
matching_tests_first=None,
2720
exclude_pattern=None,
2723
suite_decorators=None,
2725
"""Run a test suite for bzr selftest.
2727
:param runner_class: The class of runner to use. Must support the
2728
constructor arguments passed by run_suite which are more than standard
2730
:return: A boolean indicating success.
2732
TestCase._gather_lsprof_in_benchmarks = lsprof_timed
2737
if runner_class is None:
2738
runner_class = TextTestRunner
2741
runner = runner_class(stream=stream,
2743
verbosity=verbosity,
2744
bench_history=bench_history,
2745
list_only=list_only,
2748
runner.stop_on_failure=stop_on_failure
2749
# built in decorator factories:
2751
random_order(random_seed, runner),
2752
exclude_tests(exclude_pattern),
2754
if matching_tests_first:
2755
decorators.append(tests_first(pattern))
2757
decorators.append(filter_tests(pattern))
2758
if suite_decorators:
2759
decorators.extend(suite_decorators)
2760
for decorator in decorators:
2761
suite = decorator(suite)
2762
result = runner.run(suite)
2767
return result.wasStrictlySuccessful()
2769
return result.wasSuccessful()
2772
# A registry where get() returns a suite decorator.
2773
parallel_registry = registry.Registry()
2776
def fork_decorator(suite):
2777
concurrency = osutils.local_concurrency()
2778
if concurrency == 1:
2780
from testtools import ConcurrentTestSuite
2781
return ConcurrentTestSuite(suite, fork_for_tests)
2782
parallel_registry.register('fork', fork_decorator)
2785
def subprocess_decorator(suite):
2786
concurrency = osutils.local_concurrency()
2787
if concurrency == 1:
2789
from testtools import ConcurrentTestSuite
2790
return ConcurrentTestSuite(suite, reinvoke_for_tests)
2791
parallel_registry.register('subprocess', subprocess_decorator)
2794
def exclude_tests(exclude_pattern):
2795
"""Return a test suite decorator that excludes tests."""
2796
if exclude_pattern is None:
2797
return identity_decorator
2798
def decorator(suite):
2799
return ExcludeDecorator(suite, exclude_pattern)
2803
def filter_tests(pattern):
2805
return identity_decorator
2806
def decorator(suite):
2807
return FilterTestsDecorator(suite, pattern)
2811
def random_order(random_seed, runner):
2812
"""Return a test suite decorator factory for randomising tests order.
2814
:param random_seed: now, a string which casts to a long, or a long.
2815
:param runner: A test runner with a stream attribute to report on.
2817
if random_seed is None:
2818
return identity_decorator
2819
def decorator(suite):
2820
return RandomDecorator(suite, random_seed, runner.stream)
2824
def tests_first(pattern):
2826
return identity_decorator
2827
def decorator(suite):
2828
return TestFirstDecorator(suite, pattern)
2832
def identity_decorator(suite):
2837
class TestDecorator(TestSuite):
2838
"""A decorator for TestCase/TestSuite objects.
2840
Usually, subclasses should override __iter__(used when flattening test
2841
suites), which we do to filter, reorder, parallelise and so on, run() and
2845
def __init__(self, suite):
2846
TestSuite.__init__(self)
2849
def countTestCases(self):
2852
cases += test.countTestCases()
2859
def run(self, result):
2860
# Use iteration on self, not self._tests, to allow subclasses to hook
2863
if result.shouldStop:
2869
class ExcludeDecorator(TestDecorator):
2870
"""A decorator which excludes test matching an exclude pattern."""
2872
def __init__(self, suite, exclude_pattern):
2873
TestDecorator.__init__(self, suite)
2874
self.exclude_pattern = exclude_pattern
2875
self.excluded = False
2879
return iter(self._tests)
2880
self.excluded = True
2881
suite = exclude_tests_by_re(self, self.exclude_pattern)
2883
self.addTests(suite)
2884
return iter(self._tests)
2887
class FilterTestsDecorator(TestDecorator):
2888
"""A decorator which filters tests to those matching a pattern."""
2890
def __init__(self, suite, pattern):
2891
TestDecorator.__init__(self, suite)
2892
self.pattern = pattern
2893
self.filtered = False
2897
return iter(self._tests)
2898
self.filtered = True
2899
suite = filter_suite_by_re(self, self.pattern)
2901
self.addTests(suite)
2902
return iter(self._tests)
2905
class RandomDecorator(TestDecorator):
2906
"""A decorator which randomises the order of its tests."""
2908
def __init__(self, suite, random_seed, stream):
2909
TestDecorator.__init__(self, suite)
2910
self.random_seed = random_seed
2911
self.randomised = False
2912
self.stream = stream
2916
return iter(self._tests)
2917
self.randomised = True
2918
self.stream.writeln("Randomizing test order using seed %s\n" %
2919
(self.actual_seed()))
2920
# Initialise the random number generator.
2921
random.seed(self.actual_seed())
2922
suite = randomize_suite(self)
2924
self.addTests(suite)
2925
return iter(self._tests)
2927
def actual_seed(self):
2928
if self.random_seed == "now":
2929
# We convert the seed to a long to make it reuseable across
2930
# invocations (because the user can reenter it).
2931
self.random_seed = long(time.time())
2933
# Convert the seed to a long if we can
2935
self.random_seed = long(self.random_seed)
2938
return self.random_seed
2941
class TestFirstDecorator(TestDecorator):
2942
"""A decorator which moves named tests to the front."""
2944
def __init__(self, suite, pattern):
2945
TestDecorator.__init__(self, suite)
2946
self.pattern = pattern
2947
self.filtered = False
2951
return iter(self._tests)
2952
self.filtered = True
2953
suites = split_suite_by_re(self, self.pattern)
2955
self.addTests(suites)
2956
return iter(self._tests)
2959
def partition_tests(suite, count):
2960
"""Partition suite into count lists of tests."""
2962
tests = list(iter_suite_tests(suite))
2963
tests_per_process = int(math.ceil(float(len(tests)) / count))
2964
for block in range(count):
2965
low_test = block * tests_per_process
2966
high_test = low_test + tests_per_process
2967
process_tests = tests[low_test:high_test]
2968
result.append(process_tests)
2972
def fork_for_tests(suite):
2973
"""Take suite and start up one runner per CPU by forking()
2975
:return: An iterable of TestCase-like objects which can each have
2976
run(result) called on them to feed tests to result.
2978
concurrency = osutils.local_concurrency()
2980
from subunit import TestProtocolClient, ProtocolTestCase
2982
from subunit.test_results import AutoTimingTestResultDecorator
2984
AutoTimingTestResultDecorator = lambda x:x
2985
class TestInOtherProcess(ProtocolTestCase):
2986
# Should be in subunit, I think. RBC.
2987
def __init__(self, stream, pid):
2988
ProtocolTestCase.__init__(self, stream)
2991
def run(self, result):
2993
ProtocolTestCase.run(self, result)
2995
os.waitpid(self.pid, os.WNOHANG)
2997
test_blocks = partition_tests(suite, concurrency)
2998
for process_tests in test_blocks:
2999
process_suite = TestSuite()
3000
process_suite.addTests(process_tests)
3001
c2pread, c2pwrite = os.pipe()
3006
# Leave stderr and stdout open so we can see test noise
3007
# Close stdin so that the child goes away if it decides to
3008
# read from stdin (otherwise its a roulette to see what
3009
# child actually gets keystrokes for pdb etc).
3012
stream = os.fdopen(c2pwrite, 'wb', 1)
3013
subunit_result = AutoTimingTestResultDecorator(
3014
TestProtocolClient(stream))
3015
process_suite.run(subunit_result)
3020
stream = os.fdopen(c2pread, 'rb', 1)
3021
test = TestInOtherProcess(stream, pid)
3026
def reinvoke_for_tests(suite):
3027
"""Take suite and start up one runner per CPU using subprocess().
3029
:return: An iterable of TestCase-like objects which can each have
3030
run(result) called on them to feed tests to result.
3032
concurrency = osutils.local_concurrency()
3034
from subunit import ProtocolTestCase
3035
class TestInSubprocess(ProtocolTestCase):
3036
def __init__(self, process, name):
3037
ProtocolTestCase.__init__(self, process.stdout)
3038
self.process = process
3039
self.process.stdin.close()
3042
def run(self, result):
3044
ProtocolTestCase.run(self, result)
3047
os.unlink(self.name)
3048
# print "pid %d finished" % finished_process
3049
test_blocks = partition_tests(suite, concurrency)
3050
for process_tests in test_blocks:
3051
# ugly; currently reimplement rather than reuses TestCase methods.
3052
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
3053
if not os.path.isfile(bzr_path):
3054
# We are probably installed. Assume sys.argv is the right file
3055
bzr_path = sys.argv[0]
3056
fd, test_list_file_name = tempfile.mkstemp()
3057
test_list_file = os.fdopen(fd, 'wb', 1)
3058
for test in process_tests:
3059
test_list_file.write(test.id() + '\n')
3060
test_list_file.close()
3062
argv = [bzr_path, 'selftest', '--load-list', test_list_file_name,
3064
if '--no-plugins' in sys.argv:
3065
argv.append('--no-plugins')
3066
# stderr=STDOUT would be ideal, but until we prevent noise on
3067
# stderr it can interrupt the subunit protocol.
3068
process = Popen(argv, stdin=PIPE, stdout=PIPE, stderr=PIPE,
3070
test = TestInSubprocess(process, test_list_file_name)
3073
os.unlink(test_list_file_name)
3078
class BZRTransformingResult(unittest.TestResult):
3080
def __init__(self, target):
3081
unittest.TestResult.__init__(self)
3082
self.result = target
3084
def startTest(self, test):
3085
self.result.startTest(test)
3087
def stopTest(self, test):
3088
self.result.stopTest(test)
3090
def addError(self, test, err):
3091
feature = self._error_looks_like('UnavailableFeature: ', err)
3092
if feature is not None:
3093
self.result.addNotSupported(test, feature)
3095
self.result.addError(test, err)
3097
def addFailure(self, test, err):
3098
known = self._error_looks_like('KnownFailure: ', err)
3099
if known is not None:
3100
self.result._addKnownFailure(test, [KnownFailure,
3101
KnownFailure(known), None])
3103
self.result.addFailure(test, err)
3105
def addSkip(self, test, reason):
3106
self.result.addSkip(test, reason)
3108
def addSuccess(self, test):
3109
self.result.addSuccess(test)
3111
def _error_looks_like(self, prefix, err):
3112
"""Deserialize exception and returns the stringify value."""
3116
if isinstance(exc, subunit.RemoteException):
3117
# stringify the exception gives access to the remote traceback
3118
# We search the last line for 'prefix'
3119
lines = str(exc).split('\n')
3120
while lines and not lines[-1]:
3123
if lines[-1].startswith(prefix):
3124
value = lines[-1][len(prefix):]
3128
# Controlled by "bzr selftest -E=..." option
3129
selftest_debug_flags = set()
3132
def selftest(verbose=False, pattern=".*", stop_on_failure=True,
3134
test_suite_factory=None,
3137
matching_tests_first=None,
3140
exclude_pattern=None,
3146
suite_decorators=None,
3148
"""Run the whole test suite under the enhanced runner"""
3149
# XXX: Very ugly way to do this...
3150
# Disable warning about old formats because we don't want it to disturb
3151
# any blackbox tests.
3152
from bzrlib import repository
3153
repository._deprecation_warning_done = True
3155
global default_transport
3156
if transport is None:
3157
transport = default_transport
3158
old_transport = default_transport
3159
default_transport = transport
3160
global selftest_debug_flags
3161
old_debug_flags = selftest_debug_flags
3162
if debug_flags is not None:
3163
selftest_debug_flags = set(debug_flags)
3165
if load_list is None:
3168
keep_only = load_test_id_list(load_list)
3169
if test_suite_factory is None:
3170
suite = test_suite(keep_only, starting_with)
3172
suite = test_suite_factory()
3173
return run_suite(suite, 'testbzr', verbose=verbose, pattern=pattern,
3174
stop_on_failure=stop_on_failure,
3175
transport=transport,
3176
lsprof_timed=lsprof_timed,
3177
bench_history=bench_history,
3178
matching_tests_first=matching_tests_first,
3179
list_only=list_only,
3180
random_seed=random_seed,
3181
exclude_pattern=exclude_pattern,
3183
runner_class=runner_class,
3184
suite_decorators=suite_decorators,
3187
default_transport = old_transport
3188
selftest_debug_flags = old_debug_flags
3191
def load_test_id_list(file_name):
3192
"""Load a test id list from a text file.
3194
The format is one test id by line. No special care is taken to impose
3195
strict rules, these test ids are used to filter the test suite so a test id
3196
that do not match an existing test will do no harm. This allows user to add
3197
comments, leave blank lines, etc.
3201
ftest = open(file_name, 'rt')
3203
if e.errno != errno.ENOENT:
3206
raise errors.NoSuchFile(file_name)
3208
for test_name in ftest.readlines():
3209
test_list.append(test_name.strip())
3214
def suite_matches_id_list(test_suite, id_list):
3215
"""Warns about tests not appearing or appearing more than once.
3217
:param test_suite: A TestSuite object.
3218
:param test_id_list: The list of test ids that should be found in
3221
:return: (absents, duplicates) absents is a list containing the test found
3222
in id_list but not in test_suite, duplicates is a list containing the
3223
test found multiple times in test_suite.
3225
When using a prefined test id list, it may occurs that some tests do not
3226
exist anymore or that some tests use the same id. This function warns the
3227
tester about potential problems in his workflow (test lists are volatile)
3228
or in the test suite itself (using the same id for several tests does not
3229
help to localize defects).
3231
# Build a dict counting id occurrences
3233
for test in iter_suite_tests(test_suite):
3235
tests[id] = tests.get(id, 0) + 1
3240
occurs = tests.get(id, 0)
3242
not_found.append(id)
3244
duplicates.append(id)
3246
return not_found, duplicates
3249
class TestIdList(object):
3250
"""Test id list to filter a test suite.
3252
Relying on the assumption that test ids are built as:
3253
<module>[.<class>.<method>][(<param>+)], <module> being in python dotted
3254
notation, this class offers methods to :
3255
- avoid building a test suite for modules not refered to in the test list,
3256
- keep only the tests listed from the module test suite.
3259
def __init__(self, test_id_list):
3260
# When a test suite needs to be filtered against us we compare test ids
3261
# for equality, so a simple dict offers a quick and simple solution.
3262
self.tests = dict().fromkeys(test_id_list, True)
3264
# While unittest.TestCase have ids like:
3265
# <module>.<class>.<method>[(<param+)],
3266
# doctest.DocTestCase can have ids like:
3269
# <module>.<function>
3270
# <module>.<class>.<method>
3272
# Since we can't predict a test class from its name only, we settle on
3273
# a simple constraint: a test id always begins with its module name.
3276
for test_id in test_id_list:
3277
parts = test_id.split('.')
3278
mod_name = parts.pop(0)
3279
modules[mod_name] = True
3281
mod_name += '.' + part
3282
modules[mod_name] = True
3283
self.modules = modules
3285
def refers_to(self, module_name):
3286
"""Is there tests for the module or one of its sub modules."""
3287
return self.modules.has_key(module_name)
3289
def includes(self, test_id):
3290
return self.tests.has_key(test_id)
3293
class TestPrefixAliasRegistry(registry.Registry):
3294
"""A registry for test prefix aliases.
3296
This helps implement shorcuts for the --starting-with selftest
3297
option. Overriding existing prefixes is not allowed but not fatal (a
3298
warning will be emitted).
3301
def register(self, key, obj, help=None, info=None,
3302
override_existing=False):
3303
"""See Registry.register.
3305
Trying to override an existing alias causes a warning to be emitted,
3306
not a fatal execption.
3309
super(TestPrefixAliasRegistry, self).register(
3310
key, obj, help=help, info=info, override_existing=False)
3312
actual = self.get(key)
3313
note('Test prefix alias %s is already used for %s, ignoring %s'
3314
% (key, actual, obj))
3316
def resolve_alias(self, id_start):
3317
"""Replace the alias by the prefix in the given string.
3319
Using an unknown prefix is an error to help catching typos.
3321
parts = id_start.split('.')
3323
parts[0] = self.get(parts[0])
3325
raise errors.BzrCommandError(
3326
'%s is not a known test prefix alias' % parts[0])
3327
return '.'.join(parts)
3330
test_prefix_alias_registry = TestPrefixAliasRegistry()
3331
"""Registry of test prefix aliases."""
3334
# This alias allows to detect typos ('bzrlin.') by making all valid test ids
3335
# appear prefixed ('bzrlib.' is "replaced" by 'bzrlib.').
3336
test_prefix_alias_registry.register('bzrlib', 'bzrlib')
3338
# Obvious higest levels prefixes, feel free to add your own via a plugin
3339
test_prefix_alias_registry.register('bd', 'bzrlib.doc')
3340
test_prefix_alias_registry.register('bu', 'bzrlib.utils')
3341
test_prefix_alias_registry.register('bt', 'bzrlib.tests')
3342
test_prefix_alias_registry.register('bb', 'bzrlib.tests.blackbox')
3343
test_prefix_alias_registry.register('bp', 'bzrlib.plugins')
3346
def test_suite(keep_only=None, starting_with=None):
3347
"""Build and return TestSuite for the whole of bzrlib.
3349
:param keep_only: A list of test ids limiting the suite returned.
3351
:param starting_with: An id limiting the suite returned to the tests
3354
This function can be replaced if you need to change the default test
3355
suite on a global basis, but it is not encouraged.
3359
'bzrlib.tests.blackbox',
3360
'bzrlib.tests.commands',
3361
'bzrlib.tests.per_branch',
3362
'bzrlib.tests.per_bzrdir',
3363
'bzrlib.tests.per_interrepository',
3364
'bzrlib.tests.per_intertree',
3365
'bzrlib.tests.per_inventory',
3366
'bzrlib.tests.per_interbranch',
3367
'bzrlib.tests.per_lock',
3368
'bzrlib.tests.per_transport',
3369
'bzrlib.tests.per_tree',
3370
'bzrlib.tests.per_repository',
3371
'bzrlib.tests.per_repository_chk',
3372
'bzrlib.tests.per_repository_reference',
3373
'bzrlib.tests.per_workingtree',
3374
'bzrlib.tests.test__annotator',
3375
'bzrlib.tests.test__chk_map',
3376
'bzrlib.tests.test__dirstate_helpers',
3377
'bzrlib.tests.test__groupcompress',
3378
'bzrlib.tests.test__known_graph',
3379
'bzrlib.tests.test__rio',
3380
'bzrlib.tests.test__walkdirs_win32',
3381
'bzrlib.tests.test_ancestry',
3382
'bzrlib.tests.test_annotate',
3383
'bzrlib.tests.test_api',
3384
'bzrlib.tests.test_atomicfile',
3385
'bzrlib.tests.test_bad_files',
3386
'bzrlib.tests.test_bencode',
3387
'bzrlib.tests.test_bisect_multi',
3388
'bzrlib.tests.test_branch',
3389
'bzrlib.tests.test_branchbuilder',
3390
'bzrlib.tests.test_btree_index',
3391
'bzrlib.tests.test_bugtracker',
3392
'bzrlib.tests.test_bundle',
3393
'bzrlib.tests.test_bzrdir',
3394
'bzrlib.tests.test__chunks_to_lines',
3395
'bzrlib.tests.test_cache_utf8',
3396
'bzrlib.tests.test_chk_map',
3397
'bzrlib.tests.test_chk_serializer',
3398
'bzrlib.tests.test_chunk_writer',
3399
'bzrlib.tests.test_clean_tree',
3400
'bzrlib.tests.test_commands',
3401
'bzrlib.tests.test_commit',
3402
'bzrlib.tests.test_commit_merge',
3403
'bzrlib.tests.test_config',
3404
'bzrlib.tests.test_conflicts',
3405
'bzrlib.tests.test_counted_lock',
3406
'bzrlib.tests.test_decorators',
3407
'bzrlib.tests.test_delta',
3408
'bzrlib.tests.test_debug',
3409
'bzrlib.tests.test_deprecated_graph',
3410
'bzrlib.tests.test_diff',
3411
'bzrlib.tests.test_directory_service',
3412
'bzrlib.tests.test_dirstate',
3413
'bzrlib.tests.test_email_message',
3414
'bzrlib.tests.test_eol_filters',
3415
'bzrlib.tests.test_errors',
3416
'bzrlib.tests.test_export',
3417
'bzrlib.tests.test_extract',
3418
'bzrlib.tests.test_fetch',
3419
'bzrlib.tests.test_fifo_cache',
3420
'bzrlib.tests.test_filters',
3421
'bzrlib.tests.test_ftp_transport',
3422
'bzrlib.tests.test_foreign',
3423
'bzrlib.tests.test_generate_docs',
3424
'bzrlib.tests.test_generate_ids',
3425
'bzrlib.tests.test_globbing',
3426
'bzrlib.tests.test_gpg',
3427
'bzrlib.tests.test_graph',
3428
'bzrlib.tests.test_groupcompress',
3429
'bzrlib.tests.test_hashcache',
3430
'bzrlib.tests.test_help',
3431
'bzrlib.tests.test_hooks',
3432
'bzrlib.tests.test_http',
3433
'bzrlib.tests.test_http_response',
3434
'bzrlib.tests.test_https_ca_bundle',
3435
'bzrlib.tests.test_identitymap',
3436
'bzrlib.tests.test_ignores',
3437
'bzrlib.tests.test_index',
3438
'bzrlib.tests.test_info',
3439
'bzrlib.tests.test_inv',
3440
'bzrlib.tests.test_inventory_delta',
3441
'bzrlib.tests.test_knit',
3442
'bzrlib.tests.test_lazy_import',
3443
'bzrlib.tests.test_lazy_regex',
3444
'bzrlib.tests.test_lockable_files',
3445
'bzrlib.tests.test_lockdir',
3446
'bzrlib.tests.test_log',
3447
'bzrlib.tests.test_lru_cache',
3448
'bzrlib.tests.test_lsprof',
3449
'bzrlib.tests.test_mail_client',
3450
'bzrlib.tests.test_memorytree',
3451
'bzrlib.tests.test_merge',
3452
'bzrlib.tests.test_merge3',
3453
'bzrlib.tests.test_merge_core',
3454
'bzrlib.tests.test_merge_directive',
3455
'bzrlib.tests.test_missing',
3456
'bzrlib.tests.test_msgeditor',
3457
'bzrlib.tests.test_multiparent',
3458
'bzrlib.tests.test_mutabletree',
3459
'bzrlib.tests.test_nonascii',
3460
'bzrlib.tests.test_options',
3461
'bzrlib.tests.test_osutils',
3462
'bzrlib.tests.test_osutils_encodings',
3463
'bzrlib.tests.test_pack',
3464
'bzrlib.tests.test_pack_repository',
3465
'bzrlib.tests.test_patch',
3466
'bzrlib.tests.test_patches',
3467
'bzrlib.tests.test_permissions',
3468
'bzrlib.tests.test_plugins',
3469
'bzrlib.tests.test_progress',
3470
'bzrlib.tests.test_read_bundle',
3471
'bzrlib.tests.test_reconcile',
3472
'bzrlib.tests.test_reconfigure',
3473
'bzrlib.tests.test_registry',
3474
'bzrlib.tests.test_remote',
3475
'bzrlib.tests.test_rename_map',
3476
'bzrlib.tests.test_repository',
3477
'bzrlib.tests.test_revert',
3478
'bzrlib.tests.test_revision',
3479
'bzrlib.tests.test_revisionspec',
3480
'bzrlib.tests.test_revisiontree',
3481
'bzrlib.tests.test_rio',
3482
'bzrlib.tests.test_rules',
3483
'bzrlib.tests.test_sampler',
3484
'bzrlib.tests.test_selftest',
3485
'bzrlib.tests.test_serializer',
3486
'bzrlib.tests.test_setup',
3487
'bzrlib.tests.test_sftp_transport',
3488
'bzrlib.tests.test_shelf',
3489
'bzrlib.tests.test_shelf_ui',
3490
'bzrlib.tests.test_smart',
3491
'bzrlib.tests.test_smart_add',
3492
'bzrlib.tests.test_smart_request',
3493
'bzrlib.tests.test_smart_transport',
3494
'bzrlib.tests.test_smtp_connection',
3495
'bzrlib.tests.test_source',
3496
'bzrlib.tests.test_ssh_transport',
3497
'bzrlib.tests.test_status',
3498
'bzrlib.tests.test_store',
3499
'bzrlib.tests.test_strace',
3500
'bzrlib.tests.test_subsume',
3501
'bzrlib.tests.test_switch',
3502
'bzrlib.tests.test_symbol_versioning',
3503
'bzrlib.tests.test_tag',
3504
'bzrlib.tests.test_testament',
3505
'bzrlib.tests.test_textfile',
3506
'bzrlib.tests.test_textmerge',
3507
'bzrlib.tests.test_timestamp',
3508
'bzrlib.tests.test_trace',
3509
'bzrlib.tests.test_transactions',
3510
'bzrlib.tests.test_transform',
3511
'bzrlib.tests.test_transport',
3512
'bzrlib.tests.test_transport_log',
3513
'bzrlib.tests.test_tree',
3514
'bzrlib.tests.test_treebuilder',
3515
'bzrlib.tests.test_tsort',
3516
'bzrlib.tests.test_tuned_gzip',
3517
'bzrlib.tests.test_ui',
3518
'bzrlib.tests.test_uncommit',
3519
'bzrlib.tests.test_upgrade',
3520
'bzrlib.tests.test_upgrade_stacked',
3521
'bzrlib.tests.test_urlutils',
3522
'bzrlib.tests.test_version',
3523
'bzrlib.tests.test_version_info',
3524
'bzrlib.tests.test_versionedfile',
3525
'bzrlib.tests.test_weave',
3526
'bzrlib.tests.test_whitebox',
3527
'bzrlib.tests.test_win32utils',
3528
'bzrlib.tests.test_workingtree',
3529
'bzrlib.tests.test_workingtree_4',
3530
'bzrlib.tests.test_wsgi',
3531
'bzrlib.tests.test_xml',
3534
loader = TestUtil.TestLoader()
3536
if keep_only is not None:
3537
id_filter = TestIdList(keep_only)
3539
starting_with = [test_prefix_alias_registry.resolve_alias(start)
3540
for start in starting_with]
3541
# We take precedence over keep_only because *at loading time* using
3542
# both options means we will load less tests for the same final result.
3543
def interesting_module(name):
3544
for start in starting_with:
3546
# Either the module name starts with the specified string
3547
name.startswith(start)
3548
# or it may contain tests starting with the specified string
3549
or start.startswith(name)
3553
loader = TestUtil.FilteredByModuleTestLoader(interesting_module)
3555
elif keep_only is not None:
3556
loader = TestUtil.FilteredByModuleTestLoader(id_filter.refers_to)
3557
def interesting_module(name):
3558
return id_filter.refers_to(name)
3561
loader = TestUtil.TestLoader()
3562
def interesting_module(name):
3563
# No filtering, all modules are interesting
3566
suite = loader.suiteClass()
3568
# modules building their suite with loadTestsFromModuleNames
3569
suite.addTest(loader.loadTestsFromModuleNames(testmod_names))
3571
modules_to_doctest = [
3573
'bzrlib.branchbuilder',
3576
'bzrlib.iterablefile',
3580
'bzrlib.symbol_versioning',
3583
'bzrlib.version_info_formats.format_custom',
3586
for mod in modules_to_doctest:
3587
if not interesting_module(mod):
3588
# No tests to keep here, move along
3591
# note that this really does mean "report only" -- doctest
3592
# still runs the rest of the examples
3593
doc_suite = doctest.DocTestSuite(mod,
3594
optionflags=doctest.REPORT_ONLY_FIRST_FAILURE)
3595
except ValueError, e:
3596
print '**failed to get doctest for: %s\n%s' % (mod, e)
3598
if len(doc_suite._tests) == 0:
3599
raise errors.BzrError("no doctests found in %s" % (mod,))
3600
suite.addTest(doc_suite)
3602
default_encoding = sys.getdefaultencoding()
3603
for name, plugin in bzrlib.plugin.plugins().items():
3604
if not interesting_module(plugin.module.__name__):
3606
plugin_suite = plugin.test_suite()
3607
# We used to catch ImportError here and turn it into just a warning,
3608
# but really if you don't have --no-plugins this should be a failure.
3609
# mbp 20080213 - see http://bugs.launchpad.net/bugs/189771
3610
if plugin_suite is None:
3611
plugin_suite = plugin.load_plugin_tests(loader)
3612
if plugin_suite is not None:
3613
suite.addTest(plugin_suite)
3614
if default_encoding != sys.getdefaultencoding():
3615
bzrlib.trace.warning(
3616
'Plugin "%s" tried to reset default encoding to: %s', name,
3617
sys.getdefaultencoding())
3619
sys.setdefaultencoding(default_encoding)
3622
suite = filter_suite_by_id_startswith(suite, starting_with)
3624
if keep_only is not None:
3625
# Now that the referred modules have loaded their tests, keep only the
3627
suite = filter_suite_by_id_list(suite, id_filter)
3628
# Do some sanity checks on the id_list filtering
3629
not_found, duplicates = suite_matches_id_list(suite, keep_only)
3631
# The tester has used both keep_only and starting_with, so he is
3632
# already aware that some tests are excluded from the list, there
3633
# is no need to tell him which.
3636
# Some tests mentioned in the list are not in the test suite. The
3637
# list may be out of date, report to the tester.
3638
for id in not_found:
3639
bzrlib.trace.warning('"%s" not found in the test suite', id)
3640
for id in duplicates:
3641
bzrlib.trace.warning('"%s" is used as an id by several tests', id)
3646
def multiply_scenarios(scenarios_left, scenarios_right):
3647
"""Multiply two sets of scenarios.
3649
:returns: the cartesian product of the two sets of scenarios, that is
3650
a scenario for every possible combination of a left scenario and a
3654
('%s,%s' % (left_name, right_name),
3655
dict(left_dict.items() + right_dict.items()))
3656
for left_name, left_dict in scenarios_left
3657
for right_name, right_dict in scenarios_right]
3660
def multiply_tests(tests, scenarios, result):
3661
"""Multiply tests_list by scenarios into result.
3663
This is the core workhorse for test parameterisation.
3665
Typically the load_tests() method for a per-implementation test suite will
3666
call multiply_tests and return the result.
3668
:param tests: The tests to parameterise.
3669
:param scenarios: The scenarios to apply: pairs of (scenario_name,
3670
scenario_param_dict).
3671
:param result: A TestSuite to add created tests to.
3673
This returns the passed in result TestSuite with the cross product of all
3674
the tests repeated once for each scenario. Each test is adapted by adding
3675
the scenario name at the end of its id(), and updating the test object's
3676
__dict__ with the scenario_param_dict.
3678
>>> import bzrlib.tests.test_sampler
3679
>>> r = multiply_tests(
3680
... bzrlib.tests.test_sampler.DemoTest('test_nothing'),
3681
... [('one', dict(param=1)),
3682
... ('two', dict(param=2))],
3684
>>> tests = list(iter_suite_tests(r))
3688
'bzrlib.tests.test_sampler.DemoTest.test_nothing(one)'
3694
for test in iter_suite_tests(tests):
3695
apply_scenarios(test, scenarios, result)
3699
def apply_scenarios(test, scenarios, result):
3700
"""Apply the scenarios in scenarios to test and add to result.
3702
:param test: The test to apply scenarios to.
3703
:param scenarios: An iterable of scenarios to apply to test.
3705
:seealso: apply_scenario
3707
for scenario in scenarios:
3708
result.addTest(apply_scenario(test, scenario))
3712
def apply_scenario(test, scenario):
3713
"""Copy test and apply scenario to it.
3715
:param test: A test to adapt.
3716
:param scenario: A tuple describing the scenarion.
3717
The first element of the tuple is the new test id.
3718
The second element is a dict containing attributes to set on the
3720
:return: The adapted test.
3722
new_id = "%s(%s)" % (test.id(), scenario[0])
3723
new_test = clone_test(test, new_id)
3724
for name, value in scenario[1].items():
3725
setattr(new_test, name, value)
3729
def clone_test(test, new_id):
3730
"""Clone a test giving it a new id.
3732
:param test: The test to clone.
3733
:param new_id: The id to assign to it.
3734
:return: The new test.
3736
from copy import deepcopy
3737
new_test = deepcopy(test)
3738
new_test.id = lambda: new_id
3742
def _rmtree_temp_dir(dirname):
3743
# If LANG=C we probably have created some bogus paths
3744
# which rmtree(unicode) will fail to delete
3745
# so make sure we are using rmtree(str) to delete everything
3746
# except on win32, where rmtree(str) will fail
3747
# since it doesn't have the property of byte-stream paths
3748
# (they are either ascii or mbcs)
3749
if sys.platform == 'win32':
3750
# make sure we are using the unicode win32 api
3751
dirname = unicode(dirname)
3753
dirname = dirname.encode(sys.getfilesystemencoding())
3755
osutils.rmtree(dirname)
3757
if sys.platform == 'win32' and e.errno == errno.EACCES:
3758
sys.stderr.write('Permission denied: '
3759
'unable to remove testing dir '
3761
% (os.path.basename(dirname), e))
3766
class Feature(object):
3767
"""An operating system Feature."""
3770
self._available = None
3772
def available(self):
3773
"""Is the feature available?
3775
:return: True if the feature is available.
3777
if self._available is None:
3778
self._available = self._probe()
3779
return self._available
3782
"""Implement this method in concrete features.
3784
:return: True if the feature is available.
3786
raise NotImplementedError
3789
if getattr(self, 'feature_name', None):
3790
return self.feature_name()
3791
return self.__class__.__name__
3794
class _SymlinkFeature(Feature):
3797
return osutils.has_symlinks()
3799
def feature_name(self):
3802
SymlinkFeature = _SymlinkFeature()
3805
class _HardlinkFeature(Feature):
3808
return osutils.has_hardlinks()
3810
def feature_name(self):
3813
HardlinkFeature = _HardlinkFeature()
3816
class _OsFifoFeature(Feature):
3819
return getattr(os, 'mkfifo', None)
3821
def feature_name(self):
3822
return 'filesystem fifos'
3824
OsFifoFeature = _OsFifoFeature()
3827
class _UnicodeFilenameFeature(Feature):
3828
"""Does the filesystem support Unicode filenames?"""
3832
# Check for character combinations unlikely to be covered by any
3833
# single non-unicode encoding. We use the characters
3834
# - greek small letter alpha (U+03B1) and
3835
# - braille pattern dots-123456 (U+283F).
3836
os.stat(u'\u03b1\u283f')
3837
except UnicodeEncodeError:
3839
except (IOError, OSError):
3840
# The filesystem allows the Unicode filename but the file doesn't
3844
# The filesystem allows the Unicode filename and the file exists,
3848
UnicodeFilenameFeature = _UnicodeFilenameFeature()
3851
def probe_unicode_in_user_encoding():
3852
"""Try to encode several unicode strings to use in unicode-aware tests.
3853
Return first successfull match.
3855
:return: (unicode value, encoded plain string value) or (None, None)
3857
possible_vals = [u'm\xb5', u'\xe1', u'\u0410']
3858
for uni_val in possible_vals:
3860
str_val = uni_val.encode(osutils.get_user_encoding())
3861
except UnicodeEncodeError:
3862
# Try a different character
3865
return uni_val, str_val
3869
def probe_bad_non_ascii(encoding):
3870
"""Try to find [bad] character with code [128..255]
3871
that cannot be decoded to unicode in some encoding.
3872
Return None if all non-ascii characters is valid
3875
for i in xrange(128, 256):
3878
char.decode(encoding)
3879
except UnicodeDecodeError:
3884
class _HTTPSServerFeature(Feature):
3885
"""Some tests want an https Server, check if one is available.
3887
Right now, the only way this is available is under python2.6 which provides
3898
def feature_name(self):
3899
return 'HTTPSServer'
3902
HTTPSServerFeature = _HTTPSServerFeature()
3905
class _UnicodeFilename(Feature):
3906
"""Does the filesystem support Unicode filenames?"""
3911
except UnicodeEncodeError:
3913
except (IOError, OSError):
3914
# The filesystem allows the Unicode filename but the file doesn't
3918
# The filesystem allows the Unicode filename and the file exists,
3922
UnicodeFilename = _UnicodeFilename()
3925
class _UTF8Filesystem(Feature):
3926
"""Is the filesystem UTF-8?"""
3929
if osutils._fs_enc.upper() in ('UTF-8', 'UTF8'):
3933
UTF8Filesystem = _UTF8Filesystem()
3936
class _CaseInsCasePresFilenameFeature(Feature):
3937
"""Is the file-system case insensitive, but case-preserving?"""
3940
fileno, name = tempfile.mkstemp(prefix='MixedCase')
3942
# first check truly case-preserving for created files, then check
3943
# case insensitive when opening existing files.
3944
name = osutils.normpath(name)
3945
base, rel = osutils.split(name)
3946
found_rel = osutils.canonical_relpath(base, name)
3947
return (found_rel == rel
3948
and os.path.isfile(name.upper())
3949
and os.path.isfile(name.lower()))
3954
def feature_name(self):
3955
return "case-insensitive case-preserving filesystem"
3957
CaseInsCasePresFilenameFeature = _CaseInsCasePresFilenameFeature()
3960
class _CaseInsensitiveFilesystemFeature(Feature):
3961
"""Check if underlying filesystem is case-insensitive but *not* case
3964
# Note that on Windows, Cygwin, MacOS etc, the file-systems are far
3965
# more likely to be case preserving, so this case is rare.
3968
if CaseInsCasePresFilenameFeature.available():
3971
if TestCaseWithMemoryTransport.TEST_ROOT is None:
3972
root = osutils.mkdtemp(prefix='testbzr-', suffix='.tmp')
3973
TestCaseWithMemoryTransport.TEST_ROOT = root
3975
root = TestCaseWithMemoryTransport.TEST_ROOT
3976
tdir = osutils.mkdtemp(prefix='case-sensitive-probe-', suffix='',
3978
name_a = osutils.pathjoin(tdir, 'a')
3979
name_A = osutils.pathjoin(tdir, 'A')
3981
result = osutils.isdir(name_A)
3982
_rmtree_temp_dir(tdir)
3985
def feature_name(self):
3986
return 'case-insensitive filesystem'
3988
CaseInsensitiveFilesystemFeature = _CaseInsensitiveFilesystemFeature()
3991
class _SubUnitFeature(Feature):
3992
"""Check if subunit is available."""
4001
def feature_name(self):
4004
SubUnitFeature = _SubUnitFeature()
4005
# Only define SubUnitBzrRunner if subunit is available.
4007
from subunit import TestProtocolClient
4009
from subunit.test_results import AutoTimingTestResultDecorator
4011
AutoTimingTestResultDecorator = lambda x:x
4012
class SubUnitBzrRunner(TextTestRunner):
4013
def run(self, test):
4014
result = AutoTimingTestResultDecorator(
4015
TestProtocolClient(self.stream))
added
removed
bzrlib/tests/__init__.py
