1
# Copyright (C) 2005-2011 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
17
"""Testing framework extensions"""
19
# NOTE: Some classes in here use camelCaseNaming() rather than
20
# underscore_naming(). That's for consistency with unittest; it's not the
21
# general style of bzrlib. Please continue that consistency when adding e.g.
22
# new assertFoo() methods.
27
from cStringIO import StringIO
51
# nb: check this before importing anything else from within it
52
_testtools_version = getattr(testtools, '__version__', ())
53
if _testtools_version < (0, 9, 5):
54
raise ImportError("need at least testtools 0.9.5: %s is %r"
55
% (testtools.__file__, _testtools_version))
56
from testtools import content
63
commands as _mod_commands,
73
plugin as _mod_plugin,
80
transport as _mod_transport,
86
# lsprof not available
88
from bzrlib.smart import client, request
89
from bzrlib.transport import (
93
from bzrlib.symbol_versioning import (
97
from bzrlib.tests import (
103
from bzrlib.ui import NullProgressView
104
from bzrlib.ui.text import TextUIFactory
105
from bzrlib.tests.features import _CompatabilityThunkFeature
107
# Mark this python module as being part of the implementation
108
# of unittest: this gives us better tracebacks where the last
109
# shown frame is the test code, not our assertXYZ.
112
default_transport = test_server.LocalURLServer
115
_unitialized_attr = object()
116
"""A sentinel needed to act as a default value in a method signature."""
119
# Subunit result codes, defined here to prevent a hard dependency on subunit.
123
# These are intentionally brought into this namespace. That way plugins, etc
124
# can just "from bzrlib.tests import TestCase, TestLoader, etc"
125
TestSuite = TestUtil.TestSuite
126
TestLoader = TestUtil.TestLoader
128
# Tests should run in a clean and clearly defined environment. The goal is to
129
# keep them isolated from the running environment as mush as possible. The test
130
# framework ensures the variables defined below are set (or deleted if the
131
# value is None) before a test is run and reset to their original value after
132
# the test is run. Generally if some code depends on an environment variable,
133
# the tests should start without this variable in the environment. There are a
134
# few exceptions but you shouldn't violate this rule lightly.
138
# bzr now uses the Win32 API and doesn't rely on APPDATA, but the
139
# tests do check our impls match APPDATA
140
'BZR_EDITOR': None, # test_msgeditor manipulates this variable
144
'BZREMAIL': None, # may still be present in the environment
145
'EMAIL': 'jrandom@example.com', # set EMAIL as bzr does not guess
146
'BZR_PROGRESS_BAR': None,
147
# This should trap leaks to ~/.bzr.log. This occurs when tests use TestCase
148
# as a base class instead of TestCaseInTempDir. Tests inheriting from
149
# TestCase should not use disk resources, BZR_LOG is one.
150
'BZR_LOG': '/you-should-use-TestCaseInTempDir-if-you-need-a-log-file',
151
'BZR_PLUGIN_PATH': None,
152
'BZR_DISABLE_PLUGINS': None,
153
'BZR_PLUGINS_AT': None,
154
'BZR_CONCURRENCY': None,
155
# Make sure that any text ui tests are consistent regardless of
156
# the environment the test case is run in; you may want tests that
157
# test other combinations. 'dumb' is a reasonable guess for tests
158
# going to a pipe or a StringIO.
164
'SSH_AUTH_SOCK': None,
174
# Nobody cares about ftp_proxy, FTP_PROXY AFAIK. So far at
175
# least. If you do (care), please update this comment
179
'BZR_REMOTE_PATH': None,
180
# Generally speaking, we don't want apport reporting on crashes in
181
# the test envirnoment unless we're specifically testing apport,
182
# so that it doesn't leak into the real system environment. We
183
# use an env var so it propagates to subprocesses.
184
'APPORT_DISABLE': '1',
188
def override_os_environ(test, env=None):
189
"""Modify os.environ keeping a copy.
191
:param test: A test instance
193
:param env: A dict containing variable definitions to be installed
196
env = isolated_environ
197
test._original_os_environ = dict([(var, value)
198
for var, value in os.environ.iteritems()])
199
for var, value in env.iteritems():
200
osutils.set_or_unset_env(var, value)
201
if var not in test._original_os_environ:
202
# The var is new, add it with a value of None, so
203
# restore_os_environ will delete it
204
test._original_os_environ[var] = None
207
def restore_os_environ(test):
208
"""Restore os.environ to its original state.
210
:param test: A test instance previously passed to override_os_environ.
212
for var, value in test._original_os_environ.iteritems():
213
# Restore the original value (or delete it if the value has been set to
214
# None in override_os_environ).
215
osutils.set_or_unset_env(var, value)
218
def _clear__type_equality_funcs(test):
219
"""Cleanup bound methods stored on TestCase instances
221
Clear the dict breaking a few (mostly) harmless cycles in the affected
222
unittests released with Python 2.6 and initial Python 2.7 versions.
224
For a few revisions between Python 2.7.1 and Python 2.7.2 that annoyingly
225
shipped in Oneiric, an object with no clear method was used, hence the
226
extra complications, see bug 809048 for details.
228
type_equality_funcs = getattr(test, "_type_equality_funcs", None)
229
if type_equality_funcs is not None:
230
tef_clear = getattr(type_equality_funcs, "clear", None)
231
if tef_clear is None:
232
tef_instance_dict = getattr(type_equality_funcs, "__dict__", None)
233
if tef_instance_dict is not None:
234
tef_clear = tef_instance_dict.clear
235
if tef_clear is not None:
239
class ExtendedTestResult(testtools.TextTestResult):
240
"""Accepts, reports and accumulates the results of running tests.
242
Compared to the unittest version this class adds support for
243
profiling, benchmarking, stopping as soon as a test fails, and
244
skipping tests. There are further-specialized subclasses for
245
different types of display.
247
When a test finishes, in whatever way, it calls one of the addSuccess,
248
addFailure or addError classes. These in turn may redirect to a more
249
specific case for the special test results supported by our extended
252
Note that just one of these objects is fed the results from many tests.
257
def __init__(self, stream, descriptions, verbosity,
261
"""Construct new TestResult.
263
:param bench_history: Optionally, a writable file object to accumulate
266
testtools.TextTestResult.__init__(self, stream)
267
if bench_history is not None:
268
from bzrlib.version import _get_bzr_source_tree
269
src_tree = _get_bzr_source_tree()
272
revision_id = src_tree.get_parent_ids()[0]
274
# XXX: if this is a brand new tree, do the same as if there
278
# XXX: If there's no branch, what should we do?
280
bench_history.write("--date %s %s\n" % (time.time(), revision_id))
281
self._bench_history = bench_history
282
self.ui = ui.ui_factory
285
self.failure_count = 0
286
self.known_failure_count = 0
288
self.not_applicable_count = 0
289
self.unsupported = {}
291
self._overall_start_time = time.time()
292
self._strict = strict
293
self._first_thread_leaker_id = None
294
self._tests_leaking_threads_count = 0
295
self._traceback_from_test = None
297
def stopTestRun(self):
300
stopTime = time.time()
301
timeTaken = stopTime - self.startTime
302
# GZ 2010-07-19: Seems testtools has no printErrors method, and though
303
# the parent class method is similar have to duplicate
304
self._show_list('ERROR', self.errors)
305
self._show_list('FAIL', self.failures)
306
self.stream.write(self.sep2)
307
self.stream.write("%s %d test%s in %.3fs\n\n" % (actionTaken,
308
run, run != 1 and "s" or "", timeTaken))
309
if not self.wasSuccessful():
310
self.stream.write("FAILED (")
311
failed, errored = map(len, (self.failures, self.errors))
313
self.stream.write("failures=%d" % failed)
315
if failed: self.stream.write(", ")
316
self.stream.write("errors=%d" % errored)
317
if self.known_failure_count:
318
if failed or errored: self.stream.write(", ")
319
self.stream.write("known_failure_count=%d" %
320
self.known_failure_count)
321
self.stream.write(")\n")
323
if self.known_failure_count:
324
self.stream.write("OK (known_failures=%d)\n" %
325
self.known_failure_count)
327
self.stream.write("OK\n")
328
if self.skip_count > 0:
329
skipped = self.skip_count
330
self.stream.write('%d test%s skipped\n' %
331
(skipped, skipped != 1 and "s" or ""))
333
for feature, count in sorted(self.unsupported.items()):
334
self.stream.write("Missing feature '%s' skipped %d tests.\n" %
337
ok = self.wasStrictlySuccessful()
339
ok = self.wasSuccessful()
340
if self._first_thread_leaker_id:
342
'%s is leaking threads among %d leaking tests.\n' % (
343
self._first_thread_leaker_id,
344
self._tests_leaking_threads_count))
345
# We don't report the main thread as an active one.
347
'%d non-main threads were left active in the end.\n'
348
% (len(self._active_threads) - 1))
350
def getDescription(self, test):
353
def _extractBenchmarkTime(self, testCase, details=None):
354
"""Add a benchmark time for the current test case."""
355
if details and 'benchtime' in details:
356
return float(''.join(details['benchtime'].iter_bytes()))
357
return getattr(testCase, "_benchtime", None)
359
def _elapsedTestTimeString(self):
360
"""Return a time string for the overall time the current test has taken."""
361
return self._formatTime(self._delta_to_float(
362
self._now() - self._start_datetime))
364
def _testTimeString(self, testCase):
365
benchmark_time = self._extractBenchmarkTime(testCase)
366
if benchmark_time is not None:
367
return self._formatTime(benchmark_time) + "*"
369
return self._elapsedTestTimeString()
371
def _formatTime(self, seconds):
372
"""Format seconds as milliseconds with leading spaces."""
373
# some benchmarks can take thousands of seconds to run, so we need 8
375
return "%8dms" % (1000 * seconds)
377
def _shortened_test_description(self, test):
379
what = re.sub(r'^bzrlib\.tests\.', '', what)
382
# GZ 2010-10-04: Cloned tests may end up harmlessly calling this method
383
# multiple times in a row, because the handler is added for
384
# each test but the container list is shared between cases.
385
# See lp:498869 lp:625574 and lp:637725 for background.
386
def _record_traceback_from_test(self, exc_info):
387
"""Store the traceback from passed exc_info tuple till"""
388
self._traceback_from_test = exc_info[2]
390
def startTest(self, test):
391
super(ExtendedTestResult, self).startTest(test)
395
self.report_test_start(test)
396
test.number = self.count
397
self._recordTestStartTime()
398
# Make testtools cases give us the real traceback on failure
399
addOnException = getattr(test, "addOnException", None)
400
if addOnException is not None:
401
addOnException(self._record_traceback_from_test)
402
# Only check for thread leaks on bzrlib derived test cases
403
if isinstance(test, TestCase):
404
test.addCleanup(self._check_leaked_threads, test)
406
def stopTest(self, test):
407
super(ExtendedTestResult, self).stopTest(test)
408
# Manually break cycles, means touching various private things but hey
409
getDetails = getattr(test, "getDetails", None)
410
if getDetails is not None:
412
_clear__type_equality_funcs(test)
413
self._traceback_from_test = None
415
def startTests(self):
416
self.report_tests_starting()
417
self._active_threads = threading.enumerate()
419
def _check_leaked_threads(self, test):
420
"""See if any threads have leaked since last call
422
A sample of live threads is stored in the _active_threads attribute,
423
when this method runs it compares the current live threads and any not
424
in the previous sample are treated as having leaked.
426
now_active_threads = set(threading.enumerate())
427
threads_leaked = now_active_threads.difference(self._active_threads)
429
self._report_thread_leak(test, threads_leaked, now_active_threads)
430
self._tests_leaking_threads_count += 1
431
if self._first_thread_leaker_id is None:
432
self._first_thread_leaker_id = test.id()
433
self._active_threads = now_active_threads
435
def _recordTestStartTime(self):
436
"""Record that a test has started."""
437
self._start_datetime = self._now()
439
def addError(self, test, err):
440
"""Tell result that test finished with an error.
442
Called from the TestCase run() method when the test
443
fails with an unexpected error.
445
self._post_mortem(self._traceback_from_test)
446
super(ExtendedTestResult, self).addError(test, err)
447
self.error_count += 1
448
self.report_error(test, err)
452
def addFailure(self, test, err):
453
"""Tell result that test failed.
455
Called from the TestCase run() method when the test
456
fails because e.g. an assert() method failed.
458
self._post_mortem(self._traceback_from_test)
459
super(ExtendedTestResult, self).addFailure(test, err)
460
self.failure_count += 1
461
self.report_failure(test, err)
465
def addSuccess(self, test, details=None):
466
"""Tell result that test completed successfully.
468
Called from the TestCase run()
470
if self._bench_history is not None:
471
benchmark_time = self._extractBenchmarkTime(test, details)
472
if benchmark_time is not None:
473
self._bench_history.write("%s %s\n" % (
474
self._formatTime(benchmark_time),
476
self.report_success(test)
477
super(ExtendedTestResult, self).addSuccess(test)
478
test._log_contents = ''
480
def addExpectedFailure(self, test, err):
481
self.known_failure_count += 1
482
self.report_known_failure(test, err)
484
def addUnexpectedSuccess(self, test, details=None):
485
"""Tell result the test unexpectedly passed, counting as a failure
487
When the minimum version of testtools required becomes 0.9.8 this
488
can be updated to use the new handling there.
490
super(ExtendedTestResult, self).addFailure(test, details=details)
491
self.failure_count += 1
492
self.report_unexpected_success(test,
493
"".join(details["reason"].iter_text()))
497
def addNotSupported(self, test, feature):
498
"""The test will not be run because of a missing feature.
500
# this can be called in two different ways: it may be that the
501
# test started running, and then raised (through requireFeature)
502
# UnavailableFeature. Alternatively this method can be called
503
# while probing for features before running the test code proper; in
504
# that case we will see startTest and stopTest, but the test will
505
# never actually run.
506
self.unsupported.setdefault(str(feature), 0)
507
self.unsupported[str(feature)] += 1
508
self.report_unsupported(test, feature)
510
def addSkip(self, test, reason):
511
"""A test has not run for 'reason'."""
513
self.report_skip(test, reason)
515
def addNotApplicable(self, test, reason):
516
self.not_applicable_count += 1
517
self.report_not_applicable(test, reason)
519
def _count_stored_tests(self):
520
"""Count of tests instances kept alive due to not succeeding"""
521
return self.error_count + self.failure_count + self.known_failure_count
523
def _post_mortem(self, tb=None):
524
"""Start a PDB post mortem session."""
525
if os.environ.get('BZR_TEST_PDB', None):
529
def progress(self, offset, whence):
530
"""The test is adjusting the count of tests to run."""
531
if whence == SUBUNIT_SEEK_SET:
532
self.num_tests = offset
533
elif whence == SUBUNIT_SEEK_CUR:
534
self.num_tests += offset
536
raise errors.BzrError("Unknown whence %r" % whence)
538
def report_tests_starting(self):
539
"""Display information before the test run begins"""
540
if getattr(sys, 'frozen', None) is None:
541
bzr_path = osutils.realpath(sys.argv[0])
543
bzr_path = sys.executable
545
'bzr selftest: %s\n' % (bzr_path,))
548
bzrlib.__path__[0],))
550
' bzr-%s python-%s %s\n' % (
551
bzrlib.version_string,
552
bzrlib._format_version_tuple(sys.version_info),
553
platform.platform(aliased=1),
555
self.stream.write('\n')
557
def report_test_start(self, test):
558
"""Display information on the test just about to be run"""
560
def _report_thread_leak(self, test, leaked_threads, active_threads):
561
"""Display information on a test that leaked one or more threads"""
562
# GZ 2010-09-09: A leak summary reported separately from the general
563
# thread debugging would be nice. Tests under subunit
564
# need something not using stream, perhaps adding a
565
# testtools details object would be fitting.
566
if 'threads' in selftest_debug_flags:
567
self.stream.write('%s is leaking, active is now %d\n' %
568
(test.id(), len(active_threads)))
570
def startTestRun(self):
571
self.startTime = time.time()
573
def report_success(self, test):
576
def wasStrictlySuccessful(self):
577
if self.unsupported or self.known_failure_count:
579
return self.wasSuccessful()
582
class TextTestResult(ExtendedTestResult):
583
"""Displays progress and results of tests in text form"""
585
def __init__(self, stream, descriptions, verbosity,
590
ExtendedTestResult.__init__(self, stream, descriptions, verbosity,
591
bench_history, strict)
592
# We no longer pass them around, but just rely on the UIFactory stack
595
warnings.warn("Passing pb to TextTestResult is deprecated")
596
self.pb = self.ui.nested_progress_bar()
597
self.pb.show_pct = False
598
self.pb.show_spinner = False
599
self.pb.show_eta = False,
600
self.pb.show_count = False
601
self.pb.show_bar = False
602
self.pb.update_latency = 0
603
self.pb.show_transport_activity = False
605
def stopTestRun(self):
606
# called when the tests that are going to run have run
609
super(TextTestResult, self).stopTestRun()
611
def report_tests_starting(self):
612
super(TextTestResult, self).report_tests_starting()
613
self.pb.update('[test 0/%d] Starting' % (self.num_tests))
615
def _progress_prefix_text(self):
616
# the longer this text, the less space we have to show the test
618
a = '[%d' % self.count # total that have been run
619
# tests skipped as known not to be relevant are not important enough
621
## if self.skip_count:
622
## a += ', %d skip' % self.skip_count
623
## if self.known_failure_count:
624
## a += '+%dX' % self.known_failure_count
626
a +='/%d' % self.num_tests
628
runtime = time.time() - self._overall_start_time
630
a += '%dm%ds' % (runtime / 60, runtime % 60)
633
total_fail_count = self.error_count + self.failure_count
635
a += ', %d failed' % total_fail_count
636
# if self.unsupported:
637
# a += ', %d missing' % len(self.unsupported)
641
def report_test_start(self, test):
643
self._progress_prefix_text()
645
+ self._shortened_test_description(test))
647
def _test_description(self, test):
648
return self._shortened_test_description(test)
650
def report_error(self, test, err):
651
self.stream.write('ERROR: %s\n %s\n' % (
652
self._test_description(test),
656
def report_failure(self, test, err):
657
self.stream.write('FAIL: %s\n %s\n' % (
658
self._test_description(test),
662
def report_known_failure(self, test, err):
665
def report_unexpected_success(self, test, reason):
666
self.stream.write('FAIL: %s\n %s: %s\n' % (
667
self._test_description(test),
668
"Unexpected success. Should have failed",
672
def report_skip(self, test, reason):
675
def report_not_applicable(self, test, reason):
678
def report_unsupported(self, test, feature):
679
"""test cannot be run because feature is missing."""
682
class VerboseTestResult(ExtendedTestResult):
683
"""Produce long output, with one line per test run plus times"""
685
def _ellipsize_to_right(self, a_string, final_width):
686
"""Truncate and pad a string, keeping the right hand side"""
687
if len(a_string) > final_width:
688
result = '...' + a_string[3-final_width:]
691
return result.ljust(final_width)
693
def report_tests_starting(self):
694
self.stream.write('running %d tests...\n' % self.num_tests)
695
super(VerboseTestResult, self).report_tests_starting()
697
def report_test_start(self, test):
698
name = self._shortened_test_description(test)
699
width = osutils.terminal_width()
700
if width is not None:
701
# width needs space for 6 char status, plus 1 for slash, plus an
702
# 11-char time string, plus a trailing blank
703
# when NUMBERED_DIRS: plus 5 chars on test number, plus 1 char on
705
self.stream.write(self._ellipsize_to_right(name, width-18))
707
self.stream.write(name)
710
def _error_summary(self, err):
712
return '%s%s' % (indent, err[1])
714
def report_error(self, test, err):
715
self.stream.write('ERROR %s\n%s\n'
716
% (self._testTimeString(test),
717
self._error_summary(err)))
719
def report_failure(self, test, err):
720
self.stream.write(' FAIL %s\n%s\n'
721
% (self._testTimeString(test),
722
self._error_summary(err)))
724
def report_known_failure(self, test, err):
725
self.stream.write('XFAIL %s\n%s\n'
726
% (self._testTimeString(test),
727
self._error_summary(err)))
729
def report_unexpected_success(self, test, reason):
730
self.stream.write(' FAIL %s\n%s: %s\n'
731
% (self._testTimeString(test),
732
"Unexpected success. Should have failed",
735
def report_success(self, test):
736
self.stream.write(' OK %s\n' % self._testTimeString(test))
737
for bench_called, stats in getattr(test, '_benchcalls', []):
738
self.stream.write('LSProf output for %s(%s, %s)\n' % bench_called)
739
stats.pprint(file=self.stream)
740
# flush the stream so that we get smooth output. This verbose mode is
741
# used to show the output in PQM.
744
def report_skip(self, test, reason):
745
self.stream.write(' SKIP %s\n%s\n'
746
% (self._testTimeString(test), reason))
748
def report_not_applicable(self, test, reason):
749
self.stream.write(' N/A %s\n %s\n'
750
% (self._testTimeString(test), reason))
752
def report_unsupported(self, test, feature):
753
"""test cannot be run because feature is missing."""
754
self.stream.write("NODEP %s\n The feature '%s' is not available.\n"
755
%(self._testTimeString(test), feature))
758
class TextTestRunner(object):
759
stop_on_failure = False
767
result_decorators=None,
769
"""Create a TextTestRunner.
771
:param result_decorators: An optional list of decorators to apply
772
to the result object being used by the runner. Decorators are
773
applied left to right - the first element in the list is the
776
# stream may know claim to know to write unicode strings, but in older
777
# pythons this goes sufficiently wrong that it is a bad idea. (
778
# specifically a built in file with encoding 'UTF-8' will still try
779
# to encode using ascii.
780
new_encoding = osutils.get_terminal_encoding()
781
codec = codecs.lookup(new_encoding)
782
if type(codec) is tuple:
786
encode = codec.encode
787
# GZ 2010-09-08: Really we don't want to be writing arbitrary bytes,
788
# so should swap to the plain codecs.StreamWriter
789
stream = osutils.UnicodeOrBytesToBytesWriter(encode, stream,
791
stream.encoding = new_encoding
793
self.descriptions = descriptions
794
self.verbosity = verbosity
795
self._bench_history = bench_history
796
self._strict = strict
797
self._result_decorators = result_decorators or []
800
"Run the given test case or test suite."
801
if self.verbosity == 1:
802
result_class = TextTestResult
803
elif self.verbosity >= 2:
804
result_class = VerboseTestResult
805
original_result = result_class(self.stream,
808
bench_history=self._bench_history,
811
# Signal to result objects that look at stop early policy to stop,
812
original_result.stop_early = self.stop_on_failure
813
result = original_result
814
for decorator in self._result_decorators:
815
result = decorator(result)
816
result.stop_early = self.stop_on_failure
817
result.startTestRun()
822
# higher level code uses our extended protocol to determine
823
# what exit code to give.
824
return original_result
827
def iter_suite_tests(suite):
828
"""Return all tests in a suite, recursing through nested suites"""
829
if isinstance(suite, unittest.TestCase):
831
elif isinstance(suite, unittest.TestSuite):
833
for r in iter_suite_tests(item):
836
raise Exception('unknown type %r for object %r'
837
% (type(suite), suite))
840
TestSkipped = testtools.testcase.TestSkipped
843
class TestNotApplicable(TestSkipped):
844
"""A test is not applicable to the situation where it was run.
846
This is only normally raised by parameterized tests, if they find that
847
the instance they're constructed upon does not support one aspect
852
# traceback._some_str fails to format exceptions that have the default
853
# __str__ which does an implicit ascii conversion. However, repr() on those
854
# objects works, for all that its not quite what the doctor may have ordered.
855
def _clever_some_str(value):
860
return repr(value).replace('\\n', '\n')
862
return '<unprintable %s object>' % type(value).__name__
864
traceback._some_str = _clever_some_str
867
# deprecated - use self.knownFailure(), or self.expectFailure.
868
KnownFailure = testtools.testcase._ExpectedFailure
871
class UnavailableFeature(Exception):
872
"""A feature required for this test was not available.
874
This can be considered a specialised form of SkippedTest.
876
The feature should be used to construct the exception.
880
class StringIOWrapper(object):
881
"""A wrapper around cStringIO which just adds an encoding attribute.
883
Internally we can check sys.stdout to see what the output encoding
884
should be. However, cStringIO has no encoding attribute that we can
885
set. So we wrap it instead.
890
def __init__(self, s=None):
892
self.__dict__['_cstring'] = StringIO(s)
894
self.__dict__['_cstring'] = StringIO()
896
def __getattr__(self, name, getattr=getattr):
897
return getattr(self.__dict__['_cstring'], name)
899
def __setattr__(self, name, val):
900
if name == 'encoding':
901
self.__dict__['encoding'] = val
903
return setattr(self._cstring, name, val)
906
class TestUIFactory(TextUIFactory):
907
"""A UI Factory for testing.
909
Hide the progress bar but emit note()s.
911
Allows get_password to be tested without real tty attached.
913
See also CannedInputUIFactory which lets you provide programmatic input in
916
# TODO: Capture progress events at the model level and allow them to be
917
# observed by tests that care.
919
# XXX: Should probably unify more with CannedInputUIFactory or a
920
# particular configuration of TextUIFactory, or otherwise have a clearer
921
# idea of how they're supposed to be different.
922
# See https://bugs.launchpad.net/bzr/+bug/408213
924
def __init__(self, stdout=None, stderr=None, stdin=None):
925
if stdin is not None:
926
# We use a StringIOWrapper to be able to test various
927
# encodings, but the user is still responsible to
928
# encode the string and to set the encoding attribute
929
# of StringIOWrapper.
930
stdin = StringIOWrapper(stdin)
931
super(TestUIFactory, self).__init__(stdin, stdout, stderr)
933
def get_non_echoed_password(self):
934
"""Get password from stdin without trying to handle the echo mode"""
935
password = self.stdin.readline()
938
if password[-1] == '\n':
939
password = password[:-1]
942
def make_progress_view(self):
943
return NullProgressView()
946
def isolated_doctest_setUp(test):
947
override_os_environ(test)
950
def isolated_doctest_tearDown(test):
951
restore_os_environ(test)
954
def IsolatedDocTestSuite(*args, **kwargs):
955
"""Overrides doctest.DocTestSuite to handle isolation.
957
The method is really a factory and users are expected to use it as such.
960
kwargs['setUp'] = isolated_doctest_setUp
961
kwargs['tearDown'] = isolated_doctest_tearDown
962
return doctest.DocTestSuite(*args, **kwargs)
965
class TestCase(testtools.TestCase):
966
"""Base class for bzr unit tests.
968
Tests that need access to disk resources should subclass
969
TestCaseInTempDir not TestCase.
971
Error and debug log messages are redirected from their usual
972
location into a temporary file, the contents of which can be
973
retrieved by _get_log(). We use a real OS file, not an in-memory object,
974
so that it can also capture file IO. When the test completes this file
975
is read into memory and removed from disk.
977
There are also convenience functions to invoke bzr's command-line
978
routine, and to build and check bzr trees.
980
In addition to the usual method of overriding tearDown(), this class also
981
allows subclasses to register cleanup functions via addCleanup, which are
982
run in order as the object is torn down. It's less likely this will be
983
accidentally overlooked.
987
# record lsprof data when performing benchmark calls.
988
_gather_lsprof_in_benchmarks = False
990
def __init__(self, methodName='testMethod'):
991
super(TestCase, self).__init__(methodName)
992
self._directory_isolation = True
993
self.exception_handlers.insert(0,
994
(UnavailableFeature, self._do_unsupported_or_skip))
995
self.exception_handlers.insert(0,
996
(TestNotApplicable, self._do_not_applicable))
999
super(TestCase, self).setUp()
1001
timeout = config.GlobalStack().get('selftest.timeout')
1003
timeout_fixture = fixtures.TimeoutFixture(timeout)
1004
timeout_fixture.setUp()
1005
self.addCleanup(timeout_fixture.cleanUp)
1007
for feature in getattr(self, '_test_needs_features', []):
1008
self.requireFeature(feature)
1009
self._cleanEnvironment()
1011
if bzrlib.global_state is not None:
1012
self.overrideAttr(bzrlib.global_state, 'cmdline_overrides',
1013
config.CommandLineStore())
1016
self._startLogFile()
1017
self._benchcalls = []
1018
self._benchtime = None
1020
self._track_transports()
1022
self._clear_debug_flags()
1023
# Isolate global verbosity level, to make sure it's reproducible
1024
# between tests. We should get rid of this altogether: bug 656694. --
1026
self.overrideAttr(bzrlib.trace, '_verbosity_level', 0)
1027
# Isolate config option expansion until its default value for bzrlib is
1028
# settled on or a the FIXME associated with _get_expand_default_value
1029
# is addressed -- vila 20110219
1030
self.overrideAttr(config, '_expand_default_value', None)
1031
self._log_files = set()
1032
# Each key in the ``_counters`` dict holds a value for a different
1033
# counter. When the test ends, addDetail() should be used to output the
1034
# counter values. This happens in install_counter_hook().
1036
if 'config_stats' in selftest_debug_flags:
1037
self._install_config_stats_hooks()
1038
# Do not use i18n for tests (unless the test reverses this)
1044
# The sys preserved stdin/stdout should allow blackbox tests debugging
1045
pdb.Pdb(stdin=sys.__stdin__, stdout=sys.__stdout__
1046
).set_trace(sys._getframe().f_back)
1048
def discardDetail(self, name):
1049
"""Extend the addDetail, getDetails api so we can remove a detail.
1051
eg. bzr always adds the 'log' detail at startup, but we don't want to
1052
include it for skipped, xfail, etc tests.
1054
It is safe to call this for a detail that doesn't exist, in case this
1055
gets called multiple times.
1057
# We cheat. details is stored in __details which means we shouldn't
1058
# touch it. but getDetails() returns the dict directly, so we can
1060
details = self.getDetails()
1064
def install_counter_hook(self, hooks, name, counter_name=None):
1065
"""Install a counting hook.
1067
Any hook can be counted as long as it doesn't need to return a value.
1069
:param hooks: Where the hook should be installed.
1071
:param name: The hook name that will be counted.
1073
:param counter_name: The counter identifier in ``_counters``, defaults
1076
_counters = self._counters # Avoid closing over self
1077
if counter_name is None:
1079
if _counters.has_key(counter_name):
1080
raise AssertionError('%s is already used as a counter name'
1082
_counters[counter_name] = 0
1083
self.addDetail(counter_name, content.Content(content.UTF8_TEXT,
1084
lambda: ['%d' % (_counters[counter_name],)]))
1085
def increment_counter(*args, **kwargs):
1086
_counters[counter_name] += 1
1087
label = 'count %s calls' % (counter_name,)
1088
hooks.install_named_hook(name, increment_counter, label)
1089
self.addCleanup(hooks.uninstall_named_hook, name, label)
1091
def _install_config_stats_hooks(self):
1092
"""Install config hooks to count hook calls.
1095
for hook_name in ('get', 'set', 'remove', 'load', 'save'):
1096
self.install_counter_hook(config.ConfigHooks, hook_name,
1097
'config.%s' % (hook_name,))
1099
# The OldConfigHooks are private and need special handling to protect
1100
# against recursive tests (tests that run other tests), so we just do
1101
# manually what registering them into _builtin_known_hooks will provide
1103
self.overrideAttr(config, 'OldConfigHooks', config._OldConfigHooks())
1104
for hook_name in ('get', 'set', 'remove', 'load', 'save'):
1105
self.install_counter_hook(config.OldConfigHooks, hook_name,
1106
'old_config.%s' % (hook_name,))
1108
def _clear_debug_flags(self):
1109
"""Prevent externally set debug flags affecting tests.
1111
Tests that want to use debug flags can just set them in the
1112
debug_flags set during setup/teardown.
1114
# Start with a copy of the current debug flags we can safely modify.
1115
self.overrideAttr(debug, 'debug_flags', set(debug.debug_flags))
1116
if 'allow_debug' not in selftest_debug_flags:
1117
debug.debug_flags.clear()
1118
if 'disable_lock_checks' not in selftest_debug_flags:
1119
debug.debug_flags.add('strict_locks')
1121
def _clear_hooks(self):
1122
# prevent hooks affecting tests
1123
known_hooks = hooks.known_hooks
1124
self._preserved_hooks = {}
1125
for key, (parent, name) in known_hooks.iter_parent_objects():
1126
current_hooks = getattr(parent, name)
1127
self._preserved_hooks[parent] = (name, current_hooks)
1128
self._preserved_lazy_hooks = hooks._lazy_hooks
1129
hooks._lazy_hooks = {}
1130
self.addCleanup(self._restoreHooks)
1131
for key, (parent, name) in known_hooks.iter_parent_objects():
1132
factory = known_hooks.get(key)
1133
setattr(parent, name, factory())
1134
# this hook should always be installed
1135
request._install_hook()
1137
def disable_directory_isolation(self):
1138
"""Turn off directory isolation checks."""
1139
self._directory_isolation = False
1141
def enable_directory_isolation(self):
1142
"""Enable directory isolation checks."""
1143
self._directory_isolation = True
1145
def _silenceUI(self):
1146
"""Turn off UI for duration of test"""
1147
# by default the UI is off; tests can turn it on if they want it.
1148
self.overrideAttr(ui, 'ui_factory', ui.SilentUIFactory())
1150
def _check_locks(self):
1151
"""Check that all lock take/release actions have been paired."""
1152
# We always check for mismatched locks. If a mismatch is found, we
1153
# fail unless -Edisable_lock_checks is supplied to selftest, in which
1154
# case we just print a warning.
1156
acquired_locks = [lock for action, lock in self._lock_actions
1157
if action == 'acquired']
1158
released_locks = [lock for action, lock in self._lock_actions
1159
if action == 'released']
1160
broken_locks = [lock for action, lock in self._lock_actions
1161
if action == 'broken']
1162
# trivially, given the tests for lock acquistion and release, if we
1163
# have as many in each list, it should be ok. Some lock tests also
1164
# break some locks on purpose and should be taken into account by
1165
# considering that breaking a lock is just a dirty way of releasing it.
1166
if len(acquired_locks) != (len(released_locks) + len(broken_locks)):
1168
'Different number of acquired and '
1169
'released or broken locks.\n'
1173
(acquired_locks, released_locks, broken_locks))
1174
if not self._lock_check_thorough:
1175
# Rather than fail, just warn
1176
print "Broken test %s: %s" % (self, message)
1180
def _track_locks(self):
1181
"""Track lock activity during tests."""
1182
self._lock_actions = []
1183
if 'disable_lock_checks' in selftest_debug_flags:
1184
self._lock_check_thorough = False
1186
self._lock_check_thorough = True
1188
self.addCleanup(self._check_locks)
1189
_mod_lock.Lock.hooks.install_named_hook('lock_acquired',
1190
self._lock_acquired, None)
1191
_mod_lock.Lock.hooks.install_named_hook('lock_released',
1192
self._lock_released, None)
1193
_mod_lock.Lock.hooks.install_named_hook('lock_broken',
1194
self._lock_broken, None)
1196
def _lock_acquired(self, result):
1197
self._lock_actions.append(('acquired', result))
1199
def _lock_released(self, result):
1200
self._lock_actions.append(('released', result))
1202
def _lock_broken(self, result):
1203
self._lock_actions.append(('broken', result))
1205
def permit_dir(self, name):
1206
"""Permit a directory to be used by this test. See permit_url."""
1207
name_transport = _mod_transport.get_transport_from_path(name)
1208
self.permit_url(name)
1209
self.permit_url(name_transport.base)
1211
def permit_url(self, url):
1212
"""Declare that url is an ok url to use in this test.
1214
Do this for memory transports, temporary test directory etc.
1216
Do not do this for the current working directory, /tmp, or any other
1217
preexisting non isolated url.
1219
if not url.endswith('/'):
1221
self._bzr_selftest_roots.append(url)
1223
def permit_source_tree_branch_repo(self):
1224
"""Permit the source tree bzr is running from to be opened.
1226
Some code such as bzrlib.version attempts to read from the bzr branch
1227
that bzr is executing from (if any). This method permits that directory
1228
to be used in the test suite.
1230
path = self.get_source_path()
1231
self.record_directory_isolation()
1234
workingtree.WorkingTree.open(path)
1235
except (errors.NotBranchError, errors.NoWorkingTree):
1236
raise TestSkipped('Needs a working tree of bzr sources')
1238
self.enable_directory_isolation()
1240
def _preopen_isolate_transport(self, transport):
1241
"""Check that all transport openings are done in the test work area."""
1242
while isinstance(transport, pathfilter.PathFilteringTransport):
1243
# Unwrap pathfiltered transports
1244
transport = transport.server.backing_transport.clone(
1245
transport._filter('.'))
1246
url = transport.base
1247
# ReadonlySmartTCPServer_for_testing decorates the backing transport
1248
# urls it is given by prepending readonly+. This is appropriate as the
1249
# client shouldn't know that the server is readonly (or not readonly).
1250
# We could register all servers twice, with readonly+ prepending, but
1251
# that makes for a long list; this is about the same but easier to
1253
if url.startswith('readonly+'):
1254
url = url[len('readonly+'):]
1255
self._preopen_isolate_url(url)
1257
def _preopen_isolate_url(self, url):
1258
if not self._directory_isolation:
1260
if self._directory_isolation == 'record':
1261
self._bzr_selftest_roots.append(url)
1263
# This prevents all transports, including e.g. sftp ones backed on disk
1264
# from working unless they are explicitly granted permission. We then
1265
# depend on the code that sets up test transports to check that they are
1266
# appropriately isolated and enable their use by calling
1267
# self.permit_transport()
1268
if not osutils.is_inside_any(self._bzr_selftest_roots, url):
1269
raise errors.BzrError("Attempt to escape test isolation: %r %r"
1270
% (url, self._bzr_selftest_roots))
1272
def record_directory_isolation(self):
1273
"""Gather accessed directories to permit later access.
1275
This is used for tests that access the branch bzr is running from.
1277
self._directory_isolation = "record"
1279
def start_server(self, transport_server, backing_server=None):
1280
"""Start transport_server for this test.
1282
This starts the server, registers a cleanup for it and permits the
1283
server's urls to be used.
1285
if backing_server is None:
1286
transport_server.start_server()
1288
transport_server.start_server(backing_server)
1289
self.addCleanup(transport_server.stop_server)
1290
# Obtain a real transport because if the server supplies a password, it
1291
# will be hidden from the base on the client side.
1292
t = _mod_transport.get_transport_from_url(transport_server.get_url())
1293
# Some transport servers effectively chroot the backing transport;
1294
# others like SFTPServer don't - users of the transport can walk up the
1295
# transport to read the entire backing transport. This wouldn't matter
1296
# except that the workdir tests are given - and that they expect the
1297
# server's url to point at - is one directory under the safety net. So
1298
# Branch operations into the transport will attempt to walk up one
1299
# directory. Chrooting all servers would avoid this but also mean that
1300
# we wouldn't be testing directly against non-root urls. Alternatively
1301
# getting the test framework to start the server with a backing server
1302
# at the actual safety net directory would work too, but this then
1303
# means that the self.get_url/self.get_transport methods would need
1304
# to transform all their results. On balance its cleaner to handle it
1305
# here, and permit a higher url when we have one of these transports.
1306
if t.base.endswith('/work/'):
1307
# we have safety net/test root/work
1308
t = t.clone('../..')
1309
elif isinstance(transport_server,
1310
test_server.SmartTCPServer_for_testing):
1311
# The smart server adds a path similar to work, which is traversed
1312
# up from by the client. But the server is chrooted - the actual
1313
# backing transport is not escaped from, and VFS requests to the
1314
# root will error (because they try to escape the chroot).
1316
while t2.base != t.base:
1319
self.permit_url(t.base)
1321
def _track_transports(self):
1322
"""Install checks for transport usage."""
1323
# TestCase has no safe place it can write to.
1324
self._bzr_selftest_roots = []
1325
# Currently the easiest way to be sure that nothing is going on is to
1326
# hook into bzr dir opening. This leaves a small window of error for
1327
# transport tests, but they are well known, and we can improve on this
1329
bzrdir.BzrDir.hooks.install_named_hook("pre_open",
1330
self._preopen_isolate_transport, "Check bzr directories are safe.")
1332
def _ndiff_strings(self, a, b):
1333
"""Return ndiff between two strings containing lines.
1335
A trailing newline is added if missing to make the strings
1337
if b and b[-1] != '\n':
1339
if a and a[-1] != '\n':
1341
difflines = difflib.ndiff(a.splitlines(True),
1343
linejunk=lambda x: False,
1344
charjunk=lambda x: False)
1345
return ''.join(difflines)
1347
def assertEqual(self, a, b, message=''):
1351
except UnicodeError, e:
1352
# If we can't compare without getting a UnicodeError, then
1353
# obviously they are different
1354
trace.mutter('UnicodeError: %s', e)
1357
raise AssertionError("%snot equal:\na = %s\nb = %s\n"
1359
pprint.pformat(a), pprint.pformat(b)))
1361
assertEquals = assertEqual
1363
def assertEqualDiff(self, a, b, message=None):
1364
"""Assert two texts are equal, if not raise an exception.
1366
This is intended for use with multi-line strings where it can
1367
be hard to find the differences by eye.
1369
# TODO: perhaps override assertEquals to call this for strings?
1373
message = "texts not equal:\n"
1375
message = 'first string is missing a final newline.\n'
1377
message = 'second string is missing a final newline.\n'
1378
raise AssertionError(message +
1379
self._ndiff_strings(a, b))
1381
def assertEqualMode(self, mode, mode_test):
1382
self.assertEqual(mode, mode_test,
1383
'mode mismatch %o != %o' % (mode, mode_test))
1385
def assertEqualStat(self, expected, actual):
1386
"""assert that expected and actual are the same stat result.
1388
:param expected: A stat result.
1389
:param actual: A stat result.
1390
:raises AssertionError: If the expected and actual stat values differ
1391
other than by atime.
1393
self.assertEqual(expected.st_size, actual.st_size,
1394
'st_size did not match')
1395
self.assertEqual(expected.st_mtime, actual.st_mtime,
1396
'st_mtime did not match')
1397
self.assertEqual(expected.st_ctime, actual.st_ctime,
1398
'st_ctime did not match')
1399
if sys.platform == 'win32':
1400
# On Win32 both 'dev' and 'ino' cannot be trusted. In python2.4 it
1401
# is 'dev' that varies, in python 2.5 (6?) it is st_ino that is
1402
# odd. We just force it to always be 0 to avoid any problems.
1403
self.assertEqual(0, expected.st_dev)
1404
self.assertEqual(0, actual.st_dev)
1405
self.assertEqual(0, expected.st_ino)
1406
self.assertEqual(0, actual.st_ino)
1408
self.assertEqual(expected.st_dev, actual.st_dev,
1409
'st_dev did not match')
1410
self.assertEqual(expected.st_ino, actual.st_ino,
1411
'st_ino did not match')
1412
self.assertEqual(expected.st_mode, actual.st_mode,
1413
'st_mode did not match')
1415
def assertLength(self, length, obj_with_len):
1416
"""Assert that obj_with_len is of length length."""
1417
if len(obj_with_len) != length:
1418
self.fail("Incorrect length: wanted %d, got %d for %r" % (
1419
length, len(obj_with_len), obj_with_len))
1421
def assertLogsError(self, exception_class, func, *args, **kwargs):
1422
"""Assert that `func(*args, **kwargs)` quietly logs a specific error.
1425
orig_log_exception_quietly = trace.log_exception_quietly
1428
orig_log_exception_quietly()
1429
captured.append(sys.exc_info()[1])
1430
trace.log_exception_quietly = capture
1431
func(*args, **kwargs)
1433
trace.log_exception_quietly = orig_log_exception_quietly
1434
self.assertLength(1, captured)
1436
self.assertIsInstance(err, exception_class)
1439
def assertPositive(self, val):
1440
"""Assert that val is greater than 0."""
1441
self.assertTrue(val > 0, 'expected a positive value, but got %s' % val)
1443
def assertNegative(self, val):
1444
"""Assert that val is less than 0."""
1445
self.assertTrue(val < 0, 'expected a negative value, but got %s' % val)
1447
def assertStartsWith(self, s, prefix):
1448
if not s.startswith(prefix):
1449
raise AssertionError('string %r does not start with %r' % (s, prefix))
1451
def assertEndsWith(self, s, suffix):
1452
"""Asserts that s ends with suffix."""
1453
if not s.endswith(suffix):
1454
raise AssertionError('string %r does not end with %r' % (s, suffix))
1456
def assertContainsRe(self, haystack, needle_re, flags=0):
1457
"""Assert that a contains something matching a regular expression."""
1458
if not re.search(needle_re, haystack, flags):
1459
if '\n' in haystack or len(haystack) > 60:
1460
# a long string, format it in a more readable way
1461
raise AssertionError(
1462
'pattern "%s" not found in\n"""\\\n%s"""\n'
1463
% (needle_re, haystack))
1465
raise AssertionError('pattern "%s" not found in "%s"'
1466
% (needle_re, haystack))
1468
def assertNotContainsRe(self, haystack, needle_re, flags=0):
1469
"""Assert that a does not match a regular expression"""
1470
if re.search(needle_re, haystack, flags):
1471
raise AssertionError('pattern "%s" found in "%s"'
1472
% (needle_re, haystack))
1474
def assertContainsString(self, haystack, needle):
1475
if haystack.find(needle) == -1:
1476
self.fail("string %r not found in '''%s'''" % (needle, haystack))
1478
def assertNotContainsString(self, haystack, needle):
1479
if haystack.find(needle) != -1:
1480
self.fail("string %r found in '''%s'''" % (needle, haystack))
1482
def assertSubset(self, sublist, superlist):
1483
"""Assert that every entry in sublist is present in superlist."""
1484
missing = set(sublist) - set(superlist)
1485
if len(missing) > 0:
1486
raise AssertionError("value(s) %r not present in container %r" %
1487
(missing, superlist))
1489
def assertListRaises(self, excClass, func, *args, **kwargs):
1490
"""Fail unless excClass is raised when the iterator from func is used.
1492
Many functions can return generators this makes sure
1493
to wrap them in a list() call to make sure the whole generator
1494
is run, and that the proper exception is raised.
1497
list(func(*args, **kwargs))
1501
if getattr(excClass,'__name__', None) is not None:
1502
excName = excClass.__name__
1504
excName = str(excClass)
1505
raise self.failureException, "%s not raised" % excName
1507
def assertRaises(self, excClass, callableObj, *args, **kwargs):
1508
"""Assert that a callable raises a particular exception.
1510
:param excClass: As for the except statement, this may be either an
1511
exception class, or a tuple of classes.
1512
:param callableObj: A callable, will be passed ``*args`` and
1515
Returns the exception so that you can examine it.
1518
callableObj(*args, **kwargs)
1522
if getattr(excClass,'__name__', None) is not None:
1523
excName = excClass.__name__
1526
excName = str(excClass)
1527
raise self.failureException, "%s not raised" % excName
1529
def assertIs(self, left, right, message=None):
1530
if not (left is right):
1531
if message is not None:
1532
raise AssertionError(message)
1534
raise AssertionError("%r is not %r." % (left, right))
1536
def assertIsNot(self, left, right, message=None):
1538
if message is not None:
1539
raise AssertionError(message)
1541
raise AssertionError("%r is %r." % (left, right))
1543
def assertTransportMode(self, transport, path, mode):
1544
"""Fail if a path does not have mode "mode".
1546
If modes are not supported on this transport, the assertion is ignored.
1548
if not transport._can_roundtrip_unix_modebits():
1550
path_stat = transport.stat(path)
1551
actual_mode = stat.S_IMODE(path_stat.st_mode)
1552
self.assertEqual(mode, actual_mode,
1553
'mode of %r incorrect (%s != %s)'
1554
% (path, oct(mode), oct(actual_mode)))
1556
def assertIsSameRealPath(self, path1, path2):
1557
"""Fail if path1 and path2 points to different files"""
1558
self.assertEqual(osutils.realpath(path1),
1559
osutils.realpath(path2),
1560
"apparent paths:\na = %s\nb = %s\n," % (path1, path2))
1562
def assertIsInstance(self, obj, kls, msg=None):
1563
"""Fail if obj is not an instance of kls
1565
:param msg: Supplementary message to show if the assertion fails.
1567
if not isinstance(obj, kls):
1568
m = "%r is an instance of %s rather than %s" % (
1569
obj, obj.__class__, kls)
1574
def assertFileEqual(self, content, path):
1575
"""Fail if path does not contain 'content'."""
1576
self.assertPathExists(path)
1577
f = file(path, 'rb')
1582
self.assertEqualDiff(content, s)
1584
def assertDocstring(self, expected_docstring, obj):
1585
"""Fail if obj does not have expected_docstring"""
1587
# With -OO the docstring should be None instead
1588
self.assertIs(obj.__doc__, None)
1590
self.assertEqual(expected_docstring, obj.__doc__)
1592
@symbol_versioning.deprecated_method(symbol_versioning.deprecated_in((2, 4)))
1593
def failUnlessExists(self, path):
1594
return self.assertPathExists(path)
1596
def assertPathExists(self, path):
1597
"""Fail unless path or paths, which may be abs or relative, exist."""
1598
if not isinstance(path, basestring):
1600
self.assertPathExists(p)
1602
self.assertTrue(osutils.lexists(path),
1603
path + " does not exist")
1605
@symbol_versioning.deprecated_method(symbol_versioning.deprecated_in((2, 4)))
1606
def failIfExists(self, path):
1607
return self.assertPathDoesNotExist(path)
1609
def assertPathDoesNotExist(self, path):
1610
"""Fail if path or paths, which may be abs or relative, exist."""
1611
if not isinstance(path, basestring):
1613
self.assertPathDoesNotExist(p)
1615
self.assertFalse(osutils.lexists(path),
1618
def _capture_deprecation_warnings(self, a_callable, *args, **kwargs):
1619
"""A helper for callDeprecated and applyDeprecated.
1621
:param a_callable: A callable to call.
1622
:param args: The positional arguments for the callable
1623
:param kwargs: The keyword arguments for the callable
1624
:return: A tuple (warnings, result). result is the result of calling
1625
a_callable(``*args``, ``**kwargs``).
1628
def capture_warnings(msg, cls=None, stacklevel=None):
1629
# we've hooked into a deprecation specific callpath,
1630
# only deprecations should getting sent via it.
1631
self.assertEqual(cls, DeprecationWarning)
1632
local_warnings.append(msg)
1633
original_warning_method = symbol_versioning.warn
1634
symbol_versioning.set_warning_method(capture_warnings)
1636
result = a_callable(*args, **kwargs)
1638
symbol_versioning.set_warning_method(original_warning_method)
1639
return (local_warnings, result)
1641
def applyDeprecated(self, deprecation_format, a_callable, *args, **kwargs):
1642
"""Call a deprecated callable without warning the user.
1644
Note that this only captures warnings raised by symbol_versioning.warn,
1645
not other callers that go direct to the warning module.
1647
To test that a deprecated method raises an error, do something like
1648
this (remember that both assertRaises and applyDeprecated delays *args
1649
and **kwargs passing)::
1651
self.assertRaises(errors.ReservedId,
1652
self.applyDeprecated,
1653
deprecated_in((1, 5, 0)),
1657
:param deprecation_format: The deprecation format that the callable
1658
should have been deprecated with. This is the same type as the
1659
parameter to deprecated_method/deprecated_function. If the
1660
callable is not deprecated with this format, an assertion error
1662
:param a_callable: A callable to call. This may be a bound method or
1663
a regular function. It will be called with ``*args`` and
1665
:param args: The positional arguments for the callable
1666
:param kwargs: The keyword arguments for the callable
1667
:return: The result of a_callable(``*args``, ``**kwargs``)
1669
call_warnings, result = self._capture_deprecation_warnings(a_callable,
1671
expected_first_warning = symbol_versioning.deprecation_string(
1672
a_callable, deprecation_format)
1673
if len(call_warnings) == 0:
1674
self.fail("No deprecation warning generated by call to %s" %
1676
self.assertEqual(expected_first_warning, call_warnings[0])
1679
def callCatchWarnings(self, fn, *args, **kw):
1680
"""Call a callable that raises python warnings.
1682
The caller's responsible for examining the returned warnings.
1684
If the callable raises an exception, the exception is not
1685
caught and propagates up to the caller. In that case, the list
1686
of warnings is not available.
1688
:returns: ([warning_object, ...], fn_result)
1690
# XXX: This is not perfect, because it completely overrides the
1691
# warnings filters, and some code may depend on suppressing particular
1692
# warnings. It's the easiest way to insulate ourselves from -Werror,
1693
# though. -- Andrew, 20071062
1695
def _catcher(message, category, filename, lineno, file=None, line=None):
1696
# despite the name, 'message' is normally(?) a Warning subclass
1698
wlist.append(message)
1699
saved_showwarning = warnings.showwarning
1700
saved_filters = warnings.filters
1702
warnings.showwarning = _catcher
1703
warnings.filters = []
1704
result = fn(*args, **kw)
1706
warnings.showwarning = saved_showwarning
1707
warnings.filters = saved_filters
1708
return wlist, result
1710
def callDeprecated(self, expected, callable, *args, **kwargs):
1711
"""Assert that a callable is deprecated in a particular way.
1713
This is a very precise test for unusual requirements. The
1714
applyDeprecated helper function is probably more suited for most tests
1715
as it allows you to simply specify the deprecation format being used
1716
and will ensure that that is issued for the function being called.
1718
Note that this only captures warnings raised by symbol_versioning.warn,
1719
not other callers that go direct to the warning module. To catch
1720
general warnings, use callCatchWarnings.
1722
:param expected: a list of the deprecation warnings expected, in order
1723
:param callable: The callable to call
1724
:param args: The positional arguments for the callable
1725
:param kwargs: The keyword arguments for the callable
1727
call_warnings, result = self._capture_deprecation_warnings(callable,
1729
self.assertEqual(expected, call_warnings)
1732
def _startLogFile(self):
1733
"""Setup a in-memory target for bzr and testcase log messages"""
1734
pseudo_log_file = StringIO()
1735
def _get_log_contents_for_weird_testtools_api():
1736
return [pseudo_log_file.getvalue().decode(
1737
"utf-8", "replace").encode("utf-8")]
1738
self.addDetail("log", content.Content(content.ContentType("text",
1739
"plain", {"charset": "utf8"}),
1740
_get_log_contents_for_weird_testtools_api))
1741
self._log_file = pseudo_log_file
1742
self._log_memento = trace.push_log_file(self._log_file)
1743
self.addCleanup(self._finishLogFile)
1745
def _finishLogFile(self):
1746
"""Flush and dereference the in-memory log for this testcase"""
1747
if trace._trace_file:
1748
# flush the log file, to get all content
1749
trace._trace_file.flush()
1750
trace.pop_log_file(self._log_memento)
1751
# The logging module now tracks references for cleanup so discard ours
1752
del self._log_memento
1754
def thisFailsStrictLockCheck(self):
1755
"""It is known that this test would fail with -Dstrict_locks.
1757
By default, all tests are run with strict lock checking unless
1758
-Edisable_lock_checks is supplied. However there are some tests which
1759
we know fail strict locks at this point that have not been fixed.
1760
They should call this function to disable the strict checking.
1762
This should be used sparingly, it is much better to fix the locking
1763
issues rather than papering over the problem by calling this function.
1765
debug.debug_flags.discard('strict_locks')
1767
def overrideAttr(self, obj, attr_name, new=_unitialized_attr):
1768
"""Overrides an object attribute restoring it after the test.
1770
:note: This should be used with discretion; you should think about
1771
whether it's better to make the code testable without monkey-patching.
1773
:param obj: The object that will be mutated.
1775
:param attr_name: The attribute name we want to preserve/override in
1778
:param new: The optional value we want to set the attribute to.
1780
:returns: The actual attr value.
1782
value = getattr(obj, attr_name)
1783
# The actual value is captured by the call below
1784
self.addCleanup(setattr, obj, attr_name, value)
1785
if new is not _unitialized_attr:
1786
setattr(obj, attr_name, new)
1789
def overrideEnv(self, name, new):
1790
"""Set an environment variable, and reset it after the test.
1792
:param name: The environment variable name.
1794
:param new: The value to set the variable to. If None, the
1795
variable is deleted from the environment.
1797
:returns: The actual variable value.
1799
value = osutils.set_or_unset_env(name, new)
1800
self.addCleanup(osutils.set_or_unset_env, name, value)
1803
def recordCalls(self, obj, attr_name):
1804
"""Monkeypatch in a wrapper that will record calls.
1806
The monkeypatch is automatically removed when the test concludes.
1808
:param obj: The namespace holding the reference to be replaced;
1809
typically a module, class, or object.
1810
:param attr_name: A string for the name of the attribute to
1812
:returns: A list that will be extended with one item every time the
1813
function is called, with a tuple of (args, kwargs).
1817
def decorator(*args, **kwargs):
1818
calls.append((args, kwargs))
1819
return orig(*args, **kwargs)
1820
orig = self.overrideAttr(obj, attr_name, decorator)
1823
def _cleanEnvironment(self):
1824
for name, value in isolated_environ.iteritems():
1825
self.overrideEnv(name, value)
1827
def _restoreHooks(self):
1828
for klass, (name, hooks) in self._preserved_hooks.items():
1829
setattr(klass, name, hooks)
1830
self._preserved_hooks.clear()
1831
bzrlib.hooks._lazy_hooks = self._preserved_lazy_hooks
1832
self._preserved_lazy_hooks.clear()
1834
def knownFailure(self, reason):
1835
"""Declare that this test fails for a known reason
1837
Tests that are known to fail should generally be using expectedFailure
1838
with an appropriate reverse assertion if a change could cause the test
1839
to start passing. Conversely if the test has no immediate prospect of
1840
succeeding then using skip is more suitable.
1842
When this method is called while an exception is being handled, that
1843
traceback will be used, otherwise a new exception will be thrown to
1844
provide one but won't be reported.
1846
self._add_reason(reason)
1848
exc_info = sys.exc_info()
1849
if exc_info != (None, None, None):
1850
self._report_traceback(exc_info)
1853
raise self.failureException(reason)
1854
except self.failureException:
1855
exc_info = sys.exc_info()
1856
# GZ 02-08-2011: Maybe cleanup this err.exc_info attribute too?
1857
raise testtools.testcase._ExpectedFailure(exc_info)
1861
def _suppress_log(self):
1862
"""Remove the log info from details."""
1863
self.discardDetail('log')
1865
def _do_skip(self, result, reason):
1866
self._suppress_log()
1867
addSkip = getattr(result, 'addSkip', None)
1868
if not callable(addSkip):
1869
result.addSuccess(result)
1871
addSkip(self, reason)
1874
def _do_known_failure(self, result, e):
1875
self._suppress_log()
1876
err = sys.exc_info()
1877
addExpectedFailure = getattr(result, 'addExpectedFailure', None)
1878
if addExpectedFailure is not None:
1879
addExpectedFailure(self, err)
1881
result.addSuccess(self)
1884
def _do_not_applicable(self, result, e):
1886
reason = 'No reason given'
1889
self._suppress_log ()
1890
addNotApplicable = getattr(result, 'addNotApplicable', None)
1891
if addNotApplicable is not None:
1892
result.addNotApplicable(self, reason)
1894
self._do_skip(result, reason)
1897
def _report_skip(self, result, err):
1898
"""Override the default _report_skip.
1900
We want to strip the 'log' detail. If we waint until _do_skip, it has
1901
already been formatted into the 'reason' string, and we can't pull it
1904
self._suppress_log()
1905
super(TestCase, self)._report_skip(self, result, err)
1908
def _report_expected_failure(self, result, err):
1911
See _report_skip for motivation.
1913
self._suppress_log()
1914
super(TestCase, self)._report_expected_failure(self, result, err)
1917
def _do_unsupported_or_skip(self, result, e):
1919
self._suppress_log()
1920
addNotSupported = getattr(result, 'addNotSupported', None)
1921
if addNotSupported is not None:
1922
result.addNotSupported(self, reason)
1924
self._do_skip(result, reason)
1926
def time(self, callable, *args, **kwargs):
1927
"""Run callable and accrue the time it takes to the benchmark time.
1929
If lsprofiling is enabled (i.e. by --lsprof-time to bzr selftest) then
1930
this will cause lsprofile statistics to be gathered and stored in
1933
if self._benchtime is None:
1934
self.addDetail('benchtime', content.Content(content.ContentType(
1935
"text", "plain"), lambda:[str(self._benchtime)]))
1939
if not self._gather_lsprof_in_benchmarks:
1940
return callable(*args, **kwargs)
1942
# record this benchmark
1943
ret, stats = bzrlib.lsprof.profile(callable, *args, **kwargs)
1945
self._benchcalls.append(((callable, args, kwargs), stats))
1948
self._benchtime += time.time() - start
1950
def log(self, *args):
1954
"""Get a unicode string containing the log from bzrlib.trace.
1956
Undecodable characters are replaced.
1958
return u"".join(self.getDetails()['log'].iter_text())
1960
def requireFeature(self, feature):
1961
"""This test requires a specific feature is available.
1963
:raises UnavailableFeature: When feature is not available.
1965
if not feature.available():
1966
raise UnavailableFeature(feature)
1968
def _run_bzr_autosplit(self, args, retcode, encoding, stdin,
1970
"""Run bazaar command line, splitting up a string command line."""
1971
if isinstance(args, basestring):
1972
# shlex don't understand unicode strings,
1973
# so args should be plain string (bialix 20070906)
1974
args = list(shlex.split(str(args)))
1975
return self._run_bzr_core(args, retcode=retcode,
1976
encoding=encoding, stdin=stdin, working_dir=working_dir,
1979
def _run_bzr_core(self, args, retcode, encoding, stdin,
1981
# Clear chk_map page cache, because the contents are likely to mask
1983
chk_map.clear_cache()
1984
if encoding is None:
1985
encoding = osutils.get_user_encoding()
1986
stdout = StringIOWrapper()
1987
stderr = StringIOWrapper()
1988
stdout.encoding = encoding
1989
stderr.encoding = encoding
1991
self.log('run bzr: %r', args)
1992
# FIXME: don't call into logging here
1993
handler = trace.EncodedStreamHandler(stderr, errors="replace",
1995
logger = logging.getLogger('')
1996
logger.addHandler(handler)
1997
old_ui_factory = ui.ui_factory
1998
ui.ui_factory = TestUIFactory(stdin=stdin, stdout=stdout, stderr=stderr)
2001
if working_dir is not None:
2002
cwd = osutils.getcwd()
2003
os.chdir(working_dir)
2007
result = self.apply_redirected(
2008
ui.ui_factory.stdin,
2010
_mod_commands.run_bzr_catch_user_errors,
2012
except KeyboardInterrupt:
2013
# Reraise KeyboardInterrupt with contents of redirected stdout
2014
# and stderr as arguments, for tests which are interested in
2015
# stdout and stderr and are expecting the exception.
2016
out = stdout.getvalue()
2017
err = stderr.getvalue()
2019
self.log('output:\n%r', out)
2021
self.log('errors:\n%r', err)
2022
raise KeyboardInterrupt(out, err)
2024
logger.removeHandler(handler)
2025
ui.ui_factory = old_ui_factory
2029
out = stdout.getvalue()
2030
err = stderr.getvalue()
2032
self.log('output:\n%r', out)
2034
self.log('errors:\n%r', err)
2035
if retcode is not None:
2036
self.assertEquals(retcode, result,
2037
message='Unexpected return code')
2038
return result, out, err
2040
def run_bzr(self, args, retcode=0, encoding=None, stdin=None,
2041
working_dir=None, error_regexes=[], output_encoding=None):
2042
"""Invoke bzr, as if it were run from the command line.
2044
The argument list should not include the bzr program name - the
2045
first argument is normally the bzr command. Arguments may be
2046
passed in three ways:
2048
1- A list of strings, eg ["commit", "a"]. This is recommended
2049
when the command contains whitespace or metacharacters, or
2050
is built up at run time.
2052
2- A single string, eg "add a". This is the most convenient
2053
for hardcoded commands.
2055
This runs bzr through the interface that catches and reports
2056
errors, and with logging set to something approximating the
2057
default, so that error reporting can be checked.
2059
This should be the main method for tests that want to exercise the
2060
overall behavior of the bzr application (rather than a unit test
2061
or a functional test of the library.)
2063
This sends the stdout/stderr results into the test's log,
2064
where it may be useful for debugging. See also run_captured.
2066
:keyword stdin: A string to be used as stdin for the command.
2067
:keyword retcode: The status code the command should return;
2069
:keyword working_dir: The directory to run the command in
2070
:keyword error_regexes: A list of expected error messages. If
2071
specified they must be seen in the error output of the command.
2073
retcode, out, err = self._run_bzr_autosplit(
2078
working_dir=working_dir,
2080
self.assertIsInstance(error_regexes, (list, tuple))
2081
for regex in error_regexes:
2082
self.assertContainsRe(err, regex)
2085
def run_bzr_error(self, error_regexes, *args, **kwargs):
2086
"""Run bzr, and check that stderr contains the supplied regexes
2088
:param error_regexes: Sequence of regular expressions which
2089
must each be found in the error output. The relative ordering
2091
:param args: command-line arguments for bzr
2092
:param kwargs: Keyword arguments which are interpreted by run_bzr
2093
This function changes the default value of retcode to be 3,
2094
since in most cases this is run when you expect bzr to fail.
2096
:return: (out, err) The actual output of running the command (in case
2097
you want to do more inspection)
2101
# Make sure that commit is failing because there is nothing to do
2102
self.run_bzr_error(['no changes to commit'],
2103
['commit', '-m', 'my commit comment'])
2104
# Make sure --strict is handling an unknown file, rather than
2105
# giving us the 'nothing to do' error
2106
self.build_tree(['unknown'])
2107
self.run_bzr_error(['Commit refused because there are unknown files'],
2108
['commit', --strict', '-m', 'my commit comment'])
2110
kwargs.setdefault('retcode', 3)
2111
kwargs['error_regexes'] = error_regexes
2112
out, err = self.run_bzr(*args, **kwargs)
2115
def run_bzr_subprocess(self, *args, **kwargs):
2116
"""Run bzr in a subprocess for testing.
2118
This starts a new Python interpreter and runs bzr in there.
2119
This should only be used for tests that have a justifiable need for
2120
this isolation: e.g. they are testing startup time, or signal
2121
handling, or early startup code, etc. Subprocess code can't be
2122
profiled or debugged so easily.
2124
:keyword retcode: The status code that is expected. Defaults to 0. If
2125
None is supplied, the status code is not checked.
2126
:keyword env_changes: A dictionary which lists changes to environment
2127
variables. A value of None will unset the env variable.
2128
The values must be strings. The change will only occur in the
2129
child, so you don't need to fix the environment after running.
2130
:keyword universal_newlines: Convert CRLF => LF
2131
:keyword allow_plugins: By default the subprocess is run with
2132
--no-plugins to ensure test reproducibility. Also, it is possible
2133
for system-wide plugins to create unexpected output on stderr,
2134
which can cause unnecessary test failures.
2136
env_changes = kwargs.get('env_changes', {})
2137
working_dir = kwargs.get('working_dir', None)
2138
allow_plugins = kwargs.get('allow_plugins', False)
2140
if isinstance(args[0], list):
2142
elif isinstance(args[0], basestring):
2143
args = list(shlex.split(args[0]))
2145
raise ValueError("passing varargs to run_bzr_subprocess")
2146
process = self.start_bzr_subprocess(args, env_changes=env_changes,
2147
working_dir=working_dir,
2148
allow_plugins=allow_plugins)
2149
# We distinguish between retcode=None and retcode not passed.
2150
supplied_retcode = kwargs.get('retcode', 0)
2151
return self.finish_bzr_subprocess(process, retcode=supplied_retcode,
2152
universal_newlines=kwargs.get('universal_newlines', False),
2155
def start_bzr_subprocess(self, process_args, env_changes=None,
2156
skip_if_plan_to_signal=False,
2158
allow_plugins=False, stderr=subprocess.PIPE):
2159
"""Start bzr in a subprocess for testing.
2161
This starts a new Python interpreter and runs bzr in there.
2162
This should only be used for tests that have a justifiable need for
2163
this isolation: e.g. they are testing startup time, or signal
2164
handling, or early startup code, etc. Subprocess code can't be
2165
profiled or debugged so easily.
2167
:param process_args: a list of arguments to pass to the bzr executable,
2168
for example ``['--version']``.
2169
:param env_changes: A dictionary which lists changes to environment
2170
variables. A value of None will unset the env variable.
2171
The values must be strings. The change will only occur in the
2172
child, so you don't need to fix the environment after running.
2173
:param skip_if_plan_to_signal: raise TestSkipped when true and system
2174
doesn't support signalling subprocesses.
2175
:param allow_plugins: If False (default) pass --no-plugins to bzr.
2176
:param stderr: file to use for the subprocess's stderr. Valid values
2177
are those valid for the stderr argument of `subprocess.Popen`.
2178
Default value is ``subprocess.PIPE``.
2180
:returns: Popen object for the started process.
2182
if skip_if_plan_to_signal:
2183
if os.name != "posix":
2184
raise TestSkipped("Sending signals not supported")
2186
if env_changes is None:
2188
# Because $HOME is set to a tempdir for the context of a test, modules
2189
# installed in the user dir will not be found unless $PYTHONUSERBASE
2190
# gets set to the computed directory of this parent process.
2191
if site.USER_BASE is not None:
2192
env_changes["PYTHONUSERBASE"] = site.USER_BASE
2195
def cleanup_environment():
2196
for env_var, value in env_changes.iteritems():
2197
old_env[env_var] = osutils.set_or_unset_env(env_var, value)
2199
def restore_environment():
2200
for env_var, value in old_env.iteritems():
2201
osutils.set_or_unset_env(env_var, value)
2203
bzr_path = self.get_bzr_path()
2206
if working_dir is not None:
2207
cwd = osutils.getcwd()
2208
os.chdir(working_dir)
2211
# win32 subprocess doesn't support preexec_fn
2212
# so we will avoid using it on all platforms, just to
2213
# make sure the code path is used, and we don't break on win32
2214
cleanup_environment()
2215
# Include the subprocess's log file in the test details, in case
2216
# the test fails due to an error in the subprocess.
2217
self._add_subprocess_log(trace._get_bzr_log_filename())
2218
command = [sys.executable]
2219
# frozen executables don't need the path to bzr
2220
if getattr(sys, "frozen", None) is None:
2221
command.append(bzr_path)
2222
if not allow_plugins:
2223
command.append('--no-plugins')
2224
command.extend(process_args)
2225
process = self._popen(command, stdin=subprocess.PIPE,
2226
stdout=subprocess.PIPE,
2229
restore_environment()
2235
def _add_subprocess_log(self, log_file_path):
2236
if len(self._log_files) == 0:
2237
# Register an addCleanup func. We do this on the first call to
2238
# _add_subprocess_log rather than in TestCase.setUp so that this
2239
# addCleanup is registered after any cleanups for tempdirs that
2240
# subclasses might create, which will probably remove the log file
2242
self.addCleanup(self._subprocess_log_cleanup)
2243
# self._log_files is a set, so if a log file is reused we won't grab it
2245
self._log_files.add(log_file_path)
2247
def _subprocess_log_cleanup(self):
2248
for count, log_file_path in enumerate(self._log_files):
2249
# We use buffer_now=True to avoid holding the file open beyond
2250
# the life of this function, which might interfere with e.g.
2251
# cleaning tempdirs on Windows.
2252
# XXX: Testtools 0.9.5 doesn't have the content_from_file helper
2253
#detail_content = content.content_from_file(
2254
# log_file_path, buffer_now=True)
2255
with open(log_file_path, 'rb') as log_file:
2256
log_file_bytes = log_file.read()
2257
detail_content = content.Content(content.ContentType("text",
2258
"plain", {"charset": "utf8"}), lambda: [log_file_bytes])
2259
self.addDetail("start_bzr_subprocess-log-%d" % (count,),
2262
def _popen(self, *args, **kwargs):
2263
"""Place a call to Popen.
2265
Allows tests to override this method to intercept the calls made to
2266
Popen for introspection.
2268
return subprocess.Popen(*args, **kwargs)
2270
def get_source_path(self):
2271
"""Return the path of the directory containing bzrlib."""
2272
return os.path.dirname(os.path.dirname(bzrlib.__file__))
2274
def get_bzr_path(self):
2275
"""Return the path of the 'bzr' executable for this test suite."""
2276
bzr_path = os.path.join(self.get_source_path(), "bzr")
2277
if not os.path.isfile(bzr_path):
2278
# We are probably installed. Assume sys.argv is the right file
2279
bzr_path = sys.argv[0]
2282
def finish_bzr_subprocess(self, process, retcode=0, send_signal=None,
2283
universal_newlines=False, process_args=None):
2284
"""Finish the execution of process.
2286
:param process: the Popen object returned from start_bzr_subprocess.
2287
:param retcode: The status code that is expected. Defaults to 0. If
2288
None is supplied, the status code is not checked.
2289
:param send_signal: an optional signal to send to the process.
2290
:param universal_newlines: Convert CRLF => LF
2291
:returns: (stdout, stderr)
2293
if send_signal is not None:
2294
os.kill(process.pid, send_signal)
2295
out, err = process.communicate()
2297
if universal_newlines:
2298
out = out.replace('\r\n', '\n')
2299
err = err.replace('\r\n', '\n')
2301
if retcode is not None and retcode != process.returncode:
2302
if process_args is None:
2303
process_args = "(unknown args)"
2304
trace.mutter('Output of bzr %s:\n%s', process_args, out)
2305
trace.mutter('Error for bzr %s:\n%s', process_args, err)
2306
self.fail('Command bzr %s failed with retcode %s != %s'
2307
% (process_args, retcode, process.returncode))
2310
def check_tree_shape(self, tree, shape):
2311
"""Compare a tree to a list of expected names.
2313
Fail if they are not precisely equal.
2316
shape = list(shape) # copy
2317
for path, ie in tree.iter_entries_by_dir():
2318
name = path.replace('\\', '/')
2319
if ie.kind == 'directory':
2322
pass # ignore root entry
2328
self.fail("expected paths not found in inventory: %r" % shape)
2330
self.fail("unexpected paths found in inventory: %r" % extras)
2332
def apply_redirected(self, stdin=None, stdout=None, stderr=None,
2333
a_callable=None, *args, **kwargs):
2334
"""Call callable with redirected std io pipes.
2336
Returns the return code."""
2337
if not callable(a_callable):
2338
raise ValueError("a_callable must be callable.")
2340
stdin = StringIO("")
2342
if getattr(self, "_log_file", None) is not None:
2343
stdout = self._log_file
2347
if getattr(self, "_log_file", None is not None):
2348
stderr = self._log_file
2351
real_stdin = sys.stdin
2352
real_stdout = sys.stdout
2353
real_stderr = sys.stderr
2358
return a_callable(*args, **kwargs)
2360
sys.stdout = real_stdout
2361
sys.stderr = real_stderr
2362
sys.stdin = real_stdin
2364
def reduceLockdirTimeout(self):
2365
"""Reduce the default lock timeout for the duration of the test, so that
2366
if LockContention occurs during a test, it does so quickly.
2368
Tests that expect to provoke LockContention errors should call this.
2370
self.overrideAttr(lockdir, '_DEFAULT_TIMEOUT_SECONDS', 0)
2372
def make_utf8_encoded_stringio(self, encoding_type=None):
2373
"""Return a StringIOWrapper instance, that will encode Unicode
2376
if encoding_type is None:
2377
encoding_type = 'strict'
2379
output_encoding = 'utf-8'
2380
sio = codecs.getwriter(output_encoding)(sio, errors=encoding_type)
2381
sio.encoding = output_encoding
2384
def disable_verb(self, verb):
2385
"""Disable a smart server verb for one test."""
2386
from bzrlib.smart import request
2387
request_handlers = request.request_handlers
2388
orig_method = request_handlers.get(verb)
2389
orig_info = request_handlers.get_info(verb)
2390
request_handlers.remove(verb)
2391
self.addCleanup(request_handlers.register, verb, orig_method,
2395
class CapturedCall(object):
2396
"""A helper for capturing smart server calls for easy debug analysis."""
2398
def __init__(self, params, prefix_length):
2399
"""Capture the call with params and skip prefix_length stack frames."""
2402
# The last 5 frames are the __init__, the hook frame, and 3 smart
2403
# client frames. Beyond this we could get more clever, but this is good
2405
stack = traceback.extract_stack()[prefix_length:-5]
2406
self.stack = ''.join(traceback.format_list(stack))
2409
return self.call.method
2412
return self.call.method
2418
class TestCaseWithMemoryTransport(TestCase):
2419
"""Common test class for tests that do not need disk resources.
2421
Tests that need disk resources should derive from TestCaseInTempDir
2422
orTestCaseWithTransport.
2424
TestCaseWithMemoryTransport sets the TEST_ROOT variable for all bzr tests.
2426
For TestCaseWithMemoryTransport the ``test_home_dir`` is set to the name of
2427
a directory which does not exist. This serves to help ensure test isolation
2428
is preserved. ``test_dir`` is set to the TEST_ROOT, as is cwd, because they
2429
must exist. However, TestCaseWithMemoryTransport does not offer local file
2430
defaults for the transport in tests, nor does it obey the command line
2431
override, so tests that accidentally write to the common directory should
2434
:cvar TEST_ROOT: Directory containing all temporary directories, plus a
2435
``.bzr`` directory that stops us ascending higher into the filesystem.
2441
def __init__(self, methodName='runTest'):
2442
# allow test parameterization after test construction and before test
2443
# execution. Variables that the parameterizer sets need to be
2444
# ones that are not set by setUp, or setUp will trash them.
2445
super(TestCaseWithMemoryTransport, self).__init__(methodName)
2446
self.vfs_transport_factory = default_transport
2447
self.transport_server = None
2448
self.transport_readonly_server = None
2449
self.__vfs_server = None
2451
def get_transport(self, relpath=None):
2452
"""Return a writeable transport.
2454
This transport is for the test scratch space relative to
2457
:param relpath: a path relative to the base url.
2459
t = _mod_transport.get_transport_from_url(self.get_url(relpath))
2460
self.assertFalse(t.is_readonly())
2463
def get_readonly_transport(self, relpath=None):
2464
"""Return a readonly transport for the test scratch space
2466
This can be used to test that operations which should only need
2467
readonly access in fact do not try to write.
2469
:param relpath: a path relative to the base url.
2471
t = _mod_transport.get_transport_from_url(
2472
self.get_readonly_url(relpath))
2473
self.assertTrue(t.is_readonly())
2476
def create_transport_readonly_server(self):
2477
"""Create a transport server from class defined at init.
2479
This is mostly a hook for daughter classes.
2481
return self.transport_readonly_server()
2483
def get_readonly_server(self):
2484
"""Get the server instance for the readonly transport
2486
This is useful for some tests with specific servers to do diagnostics.
2488
if self.__readonly_server is None:
2489
if self.transport_readonly_server is None:
2490
# readonly decorator requested
2491
self.__readonly_server = test_server.ReadonlyServer()
2493
# explicit readonly transport.
2494
self.__readonly_server = self.create_transport_readonly_server()
2495
self.start_server(self.__readonly_server,
2496
self.get_vfs_only_server())
2497
return self.__readonly_server
2499
def get_readonly_url(self, relpath=None):
2500
"""Get a URL for the readonly transport.
2502
This will either be backed by '.' or a decorator to the transport
2503
used by self.get_url()
2504
relpath provides for clients to get a path relative to the base url.
2505
These should only be downwards relative, not upwards.
2507
base = self.get_readonly_server().get_url()
2508
return self._adjust_url(base, relpath)
2510
def get_vfs_only_server(self):
2511
"""Get the vfs only read/write server instance.
2513
This is useful for some tests with specific servers that need
2516
For TestCaseWithMemoryTransport this is always a MemoryServer, and there
2517
is no means to override it.
2519
if self.__vfs_server is None:
2520
self.__vfs_server = memory.MemoryServer()
2521
self.start_server(self.__vfs_server)
2522
return self.__vfs_server
2524
def get_server(self):
2525
"""Get the read/write server instance.
2527
This is useful for some tests with specific servers that need
2530
This is built from the self.transport_server factory. If that is None,
2531
then the self.get_vfs_server is returned.
2533
if self.__server is None:
2534
if (self.transport_server is None or self.transport_server is
2535
self.vfs_transport_factory):
2536
self.__server = self.get_vfs_only_server()
2538
# bring up a decorated means of access to the vfs only server.
2539
self.__server = self.transport_server()
2540
self.start_server(self.__server, self.get_vfs_only_server())
2541
return self.__server
2543
def _adjust_url(self, base, relpath):
2544
"""Get a URL (or maybe a path) for the readwrite transport.
2546
This will either be backed by '.' or to an equivalent non-file based
2548
relpath provides for clients to get a path relative to the base url.
2549
These should only be downwards relative, not upwards.
2551
if relpath is not None and relpath != '.':
2552
if not base.endswith('/'):
2554
# XXX: Really base should be a url; we did after all call
2555
# get_url()! But sometimes it's just a path (from
2556
# LocalAbspathServer), and it'd be wrong to append urlescaped data
2557
# to a non-escaped local path.
2558
if base.startswith('./') or base.startswith('/'):
2561
base += urlutils.escape(relpath)
2564
def get_url(self, relpath=None):
2565
"""Get a URL (or maybe a path) for the readwrite transport.
2567
This will either be backed by '.' or to an equivalent non-file based
2569
relpath provides for clients to get a path relative to the base url.
2570
These should only be downwards relative, not upwards.
2572
base = self.get_server().get_url()
2573
return self._adjust_url(base, relpath)
2575
def get_vfs_only_url(self, relpath=None):
2576
"""Get a URL (or maybe a path for the plain old vfs transport.
2578
This will never be a smart protocol. It always has all the
2579
capabilities of the local filesystem, but it might actually be a
2580
MemoryTransport or some other similar virtual filesystem.
2582
This is the backing transport (if any) of the server returned by
2583
get_url and get_readonly_url.
2585
:param relpath: provides for clients to get a path relative to the base
2586
url. These should only be downwards relative, not upwards.
2589
base = self.get_vfs_only_server().get_url()
2590
return self._adjust_url(base, relpath)
2592
def _create_safety_net(self):
2593
"""Make a fake bzr directory.
2595
This prevents any tests propagating up onto the TEST_ROOT directory's
2598
root = TestCaseWithMemoryTransport.TEST_ROOT
2599
# Make sure we get a readable and accessible home for .bzr.log
2600
# and/or config files, and not fallback to weird defaults (see
2601
# http://pad.lv/825027).
2602
self.assertIs(None, os.environ.get('BZR_HOME', None))
2603
os.environ['BZR_HOME'] = root
2604
wt = bzrdir.BzrDir.create_standalone_workingtree(root)
2605
del os.environ['BZR_HOME']
2606
# Hack for speed: remember the raw bytes of the dirstate file so that
2607
# we don't need to re-open the wt to check it hasn't changed.
2608
TestCaseWithMemoryTransport._SAFETY_NET_PRISTINE_DIRSTATE = (
2609
wt.control_transport.get_bytes('dirstate'))
2611
def _check_safety_net(self):
2612
"""Check that the safety .bzr directory have not been touched.
2614
_make_test_root have created a .bzr directory to prevent tests from
2615
propagating. This method ensures than a test did not leaked.
2617
root = TestCaseWithMemoryTransport.TEST_ROOT
2618
t = _mod_transport.get_transport_from_path(root)
2619
self.permit_url(t.base)
2620
if (t.get_bytes('.bzr/checkout/dirstate') !=
2621
TestCaseWithMemoryTransport._SAFETY_NET_PRISTINE_DIRSTATE):
2622
# The current test have modified the /bzr directory, we need to
2623
# recreate a new one or all the followng tests will fail.
2624
# If you need to inspect its content uncomment the following line
2625
# import pdb; pdb.set_trace()
2626
_rmtree_temp_dir(root + '/.bzr', test_id=self.id())
2627
self._create_safety_net()
2628
raise AssertionError('%s/.bzr should not be modified' % root)
2630
def _make_test_root(self):
2631
if TestCaseWithMemoryTransport.TEST_ROOT is None:
2632
# Watch out for tricky test dir (on OSX /tmp -> /private/tmp)
2633
root = osutils.realpath(osutils.mkdtemp(prefix='testbzr-',
2635
TestCaseWithMemoryTransport.TEST_ROOT = root
2637
self._create_safety_net()
2639
# The same directory is used by all tests, and we're not
2640
# specifically told when all tests are finished. This will do.
2641
atexit.register(_rmtree_temp_dir, root)
2643
self.permit_dir(TestCaseWithMemoryTransport.TEST_ROOT)
2644
self.addCleanup(self._check_safety_net)
2646
def makeAndChdirToTestDir(self):
2647
"""Create a temporary directories for this one test.
2649
This must set self.test_home_dir and self.test_dir and chdir to
2652
For TestCaseWithMemoryTransport we chdir to the TEST_ROOT for this test.
2654
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2655
self.test_dir = TestCaseWithMemoryTransport.TEST_ROOT
2656
self.test_home_dir = self.test_dir + "/MemoryTransportMissingHomeDir"
2657
self.permit_dir(self.test_dir)
2659
def make_branch(self, relpath, format=None):
2660
"""Create a branch on the transport at relpath."""
2661
repo = self.make_repository(relpath, format=format)
2662
return repo.bzrdir.create_branch(append_revisions_only=False)
2664
def get_default_format(self):
2667
def resolve_format(self, format):
2668
"""Resolve an object to a ControlDir format object.
2670
The initial format object can either already be
2671
a ControlDirFormat, None (for the default format),
2672
or a string with the name of the control dir format.
2674
:param format: Object to resolve
2675
:return A ControlDirFormat instance
2678
format = self.get_default_format()
2679
if isinstance(format, basestring):
2680
format = bzrdir.format_registry.make_bzrdir(format)
2683
def make_bzrdir(self, relpath, format=None):
2685
# might be a relative or absolute path
2686
maybe_a_url = self.get_url(relpath)
2687
segments = maybe_a_url.rsplit('/', 1)
2688
t = _mod_transport.get_transport(maybe_a_url)
2689
if len(segments) > 1 and segments[-1] not in ('', '.'):
2691
format = self.resolve_format(format)
2692
return format.initialize_on_transport(t)
2693
except errors.UninitializableFormat:
2694
raise TestSkipped("Format %s is not initializable." % format)
2696
def make_repository(self, relpath, shared=None, format=None):
2697
"""Create a repository on our default transport at relpath.
2699
Note that relpath must be a relative path, not a full url.
2701
# FIXME: If you create a remoterepository this returns the underlying
2702
# real format, which is incorrect. Actually we should make sure that
2703
# RemoteBzrDir returns a RemoteRepository.
2704
# maybe mbp 20070410
2705
made_control = self.make_bzrdir(relpath, format=format)
2706
return made_control.create_repository(shared=shared)
2708
def make_smart_server(self, path, backing_server=None):
2709
if backing_server is None:
2710
backing_server = self.get_server()
2711
smart_server = test_server.SmartTCPServer_for_testing()
2712
self.start_server(smart_server, backing_server)
2713
remote_transport = _mod_transport.get_transport_from_url(smart_server.get_url()
2715
return remote_transport
2717
def make_branch_and_memory_tree(self, relpath, format=None):
2718
"""Create a branch on the default transport and a MemoryTree for it."""
2719
b = self.make_branch(relpath, format=format)
2720
return memorytree.MemoryTree.create_on_branch(b)
2722
def make_branch_builder(self, relpath, format=None):
2723
branch = self.make_branch(relpath, format=format)
2724
return branchbuilder.BranchBuilder(branch=branch)
2726
def overrideEnvironmentForTesting(self):
2727
test_home_dir = self.test_home_dir
2728
if isinstance(test_home_dir, unicode):
2729
test_home_dir = test_home_dir.encode(sys.getfilesystemencoding())
2730
self.overrideEnv('HOME', test_home_dir)
2731
self.overrideEnv('BZR_HOME', test_home_dir)
2734
super(TestCaseWithMemoryTransport, self).setUp()
2735
# Ensure that ConnectedTransport doesn't leak sockets
2736
def get_transport_from_url_with_cleanup(*args, **kwargs):
2737
t = orig_get_transport_from_url(*args, **kwargs)
2738
if isinstance(t, _mod_transport.ConnectedTransport):
2739
self.addCleanup(t.disconnect)
2742
orig_get_transport_from_url = self.overrideAttr(
2743
_mod_transport, 'get_transport_from_url',
2744
get_transport_from_url_with_cleanup)
2745
self._make_test_root()
2746
self.addCleanup(os.chdir, os.getcwdu())
2747
self.makeAndChdirToTestDir()
2748
self.overrideEnvironmentForTesting()
2749
self.__readonly_server = None
2750
self.__server = None
2751
self.reduceLockdirTimeout()
2753
def setup_smart_server_with_call_log(self):
2754
"""Sets up a smart server as the transport server with a call log."""
2755
self.transport_server = test_server.SmartTCPServer_for_testing
2756
self.hpss_calls = []
2758
# Skip the current stack down to the caller of
2759
# setup_smart_server_with_call_log
2760
prefix_length = len(traceback.extract_stack()) - 2
2761
def capture_hpss_call(params):
2762
self.hpss_calls.append(
2763
CapturedCall(params, prefix_length))
2764
client._SmartClient.hooks.install_named_hook(
2765
'call', capture_hpss_call, None)
2767
def reset_smart_call_log(self):
2768
self.hpss_calls = []
2771
class TestCaseInTempDir(TestCaseWithMemoryTransport):
2772
"""Derived class that runs a test within a temporary directory.
2774
This is useful for tests that need to create a branch, etc.
2776
The directory is created in a slightly complex way: for each
2777
Python invocation, a new temporary top-level directory is created.
2778
All test cases create their own directory within that. If the
2779
tests complete successfully, the directory is removed.
2781
:ivar test_base_dir: The path of the top-level directory for this
2782
test, which contains a home directory and a work directory.
2784
:ivar test_home_dir: An initially empty directory under test_base_dir
2785
which is used as $HOME for this test.
2787
:ivar test_dir: A directory under test_base_dir used as the current
2788
directory when the test proper is run.
2791
OVERRIDE_PYTHON = 'python'
2794
super(TestCaseInTempDir, self).setUp()
2795
# Remove the protection set in isolated_environ, we have a proper
2796
# access to disk resources now.
2797
self.overrideEnv('BZR_LOG', None)
2799
def check_file_contents(self, filename, expect):
2800
self.log("check contents of file %s" % filename)
2806
if contents != expect:
2807
self.log("expected: %r" % expect)
2808
self.log("actually: %r" % contents)
2809
self.fail("contents of %s not as expected" % filename)
2811
def _getTestDirPrefix(self):
2812
# create a directory within the top level test directory
2813
if sys.platform in ('win32', 'cygwin'):
2814
name_prefix = re.sub('[<>*=+",:;_/\\-]', '_', self.id())
2815
# windows is likely to have path-length limits so use a short name
2816
name_prefix = name_prefix[-30:]
2818
name_prefix = re.sub('[/]', '_', self.id())
2821
def makeAndChdirToTestDir(self):
2822
"""See TestCaseWithMemoryTransport.makeAndChdirToTestDir().
2824
For TestCaseInTempDir we create a temporary directory based on the test
2825
name and then create two subdirs - test and home under it.
2827
name_prefix = osutils.pathjoin(TestCaseWithMemoryTransport.TEST_ROOT,
2828
self._getTestDirPrefix())
2830
for i in range(100):
2831
if os.path.exists(name):
2832
name = name_prefix + '_' + str(i)
2834
# now create test and home directories within this dir
2835
self.test_base_dir = name
2836
self.addCleanup(self.deleteTestDir)
2837
os.mkdir(self.test_base_dir)
2839
self.permit_dir(self.test_base_dir)
2840
# 'sprouting' and 'init' of a branch both walk up the tree to find
2841
# stacking policy to honour; create a bzr dir with an unshared
2842
# repository (but not a branch - our code would be trying to escape
2843
# then!) to stop them, and permit it to be read.
2844
# control = bzrdir.BzrDir.create(self.test_base_dir)
2845
# control.create_repository()
2846
self.test_home_dir = self.test_base_dir + '/home'
2847
os.mkdir(self.test_home_dir)
2848
self.test_dir = self.test_base_dir + '/work'
2849
os.mkdir(self.test_dir)
2850
os.chdir(self.test_dir)
2851
# put name of test inside
2852
f = file(self.test_base_dir + '/name', 'w')
2858
def deleteTestDir(self):
2859
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2860
_rmtree_temp_dir(self.test_base_dir, test_id=self.id())
2862
def build_tree(self, shape, line_endings='binary', transport=None):
2863
"""Build a test tree according to a pattern.
2865
shape is a sequence of file specifications. If the final
2866
character is '/', a directory is created.
2868
This assumes that all the elements in the tree being built are new.
2870
This doesn't add anything to a branch.
2872
:type shape: list or tuple.
2873
:param line_endings: Either 'binary' or 'native'
2874
in binary mode, exact contents are written in native mode, the
2875
line endings match the default platform endings.
2876
:param transport: A transport to write to, for building trees on VFS's.
2877
If the transport is readonly or None, "." is opened automatically.
2880
if type(shape) not in (list, tuple):
2881
raise AssertionError("Parameter 'shape' should be "
2882
"a list or a tuple. Got %r instead" % (shape,))
2883
# It's OK to just create them using forward slashes on windows.
2884
if transport is None or transport.is_readonly():
2885
transport = _mod_transport.get_transport_from_path(".")
2887
self.assertIsInstance(name, basestring)
2889
transport.mkdir(urlutils.escape(name[:-1]))
2891
if line_endings == 'binary':
2893
elif line_endings == 'native':
2896
raise errors.BzrError(
2897
'Invalid line ending request %r' % line_endings)
2898
content = "contents of %s%s" % (name.encode('utf-8'), end)
2899
transport.put_bytes_non_atomic(urlutils.escape(name), content)
2901
build_tree_contents = staticmethod(treeshape.build_tree_contents)
2903
def assertInWorkingTree(self, path, root_path='.', tree=None):
2904
"""Assert whether path or paths are in the WorkingTree"""
2906
tree = workingtree.WorkingTree.open(root_path)
2907
if not isinstance(path, basestring):
2909
self.assertInWorkingTree(p, tree=tree)
2911
self.assertIsNot(tree.path2id(path), None,
2912
path+' not in working tree.')
2914
def assertNotInWorkingTree(self, path, root_path='.', tree=None):
2915
"""Assert whether path or paths are not in the WorkingTree"""
2917
tree = workingtree.WorkingTree.open(root_path)
2918
if not isinstance(path, basestring):
2920
self.assertNotInWorkingTree(p,tree=tree)
2922
self.assertIs(tree.path2id(path), None, path+' in working tree.')
2925
class TestCaseWithTransport(TestCaseInTempDir):
2926
"""A test case that provides get_url and get_readonly_url facilities.
2928
These back onto two transport servers, one for readonly access and one for
2931
If no explicit class is provided for readonly access, a
2932
ReadonlyTransportDecorator is used instead which allows the use of non disk
2933
based read write transports.
2935
If an explicit class is provided for readonly access, that server and the
2936
readwrite one must both define get_url() as resolving to os.getcwd().
2939
def get_vfs_only_server(self):
2940
"""See TestCaseWithMemoryTransport.
2942
This is useful for some tests with specific servers that need
2945
if self.__vfs_server is None:
2946
self.__vfs_server = self.vfs_transport_factory()
2947
self.start_server(self.__vfs_server)
2948
return self.__vfs_server
2950
def make_branch_and_tree(self, relpath, format=None):
2951
"""Create a branch on the transport and a tree locally.
2953
If the transport is not a LocalTransport, the Tree can't be created on
2954
the transport. In that case if the vfs_transport_factory is
2955
LocalURLServer the working tree is created in the local
2956
directory backing the transport, and the returned tree's branch and
2957
repository will also be accessed locally. Otherwise a lightweight
2958
checkout is created and returned.
2960
We do this because we can't physically create a tree in the local
2961
path, with a branch reference to the transport_factory url, and
2962
a branch + repository in the vfs_transport, unless the vfs_transport
2963
namespace is distinct from the local disk - the two branch objects
2964
would collide. While we could construct a tree with its branch object
2965
pointing at the transport_factory transport in memory, reopening it
2966
would behaving unexpectedly, and has in the past caused testing bugs
2967
when we tried to do it that way.
2969
:param format: The BzrDirFormat.
2970
:returns: the WorkingTree.
2972
# TODO: always use the local disk path for the working tree,
2973
# this obviously requires a format that supports branch references
2974
# so check for that by checking bzrdir.BzrDirFormat.get_default_format()
2976
format = self.resolve_format(format=format)
2977
if not format.supports_workingtrees:
2978
b = self.make_branch(relpath+'.branch', format=format)
2979
return b.create_checkout(relpath, lightweight=True)
2980
b = self.make_branch(relpath, format=format)
2982
return b.bzrdir.create_workingtree()
2983
except errors.NotLocalUrl:
2984
# We can only make working trees locally at the moment. If the
2985
# transport can't support them, then we keep the non-disk-backed
2986
# branch and create a local checkout.
2987
if self.vfs_transport_factory is test_server.LocalURLServer:
2988
# the branch is colocated on disk, we cannot create a checkout.
2989
# hopefully callers will expect this.
2990
local_controldir= bzrdir.BzrDir.open(self.get_vfs_only_url(relpath))
2991
wt = local_controldir.create_workingtree()
2992
if wt.branch._format != b._format:
2994
# Make sure that assigning to wt._branch fixes wt.branch,
2995
# in case the implementation details of workingtree objects
2997
self.assertIs(b, wt.branch)
3000
return b.create_checkout(relpath, lightweight=True)
3002
def assertIsDirectory(self, relpath, transport):
3003
"""Assert that relpath within transport is a directory.
3005
This may not be possible on all transports; in that case it propagates
3006
a TransportNotPossible.
3009
mode = transport.stat(relpath).st_mode
3010
except errors.NoSuchFile:
3011
self.fail("path %s is not a directory; no such file"
3013
if not stat.S_ISDIR(mode):
3014
self.fail("path %s is not a directory; has mode %#o"
3017
def assertTreesEqual(self, left, right):
3018
"""Check that left and right have the same content and properties."""
3019
# we use a tree delta to check for equality of the content, and we
3020
# manually check for equality of other things such as the parents list.
3021
self.assertEqual(left.get_parent_ids(), right.get_parent_ids())
3022
differences = left.changes_from(right)
3023
self.assertFalse(differences.has_changed(),
3024
"Trees %r and %r are different: %r" % (left, right, differences))
3027
super(TestCaseWithTransport, self).setUp()
3028
self.__vfs_server = None
3030
def disable_missing_extensions_warning(self):
3031
"""Some tests expect a precise stderr content.
3033
There is no point in forcing them to duplicate the extension related
3036
config.GlobalConfig().set_user_option('ignore_missing_extensions', True)
3039
class ChrootedTestCase(TestCaseWithTransport):
3040
"""A support class that provides readonly urls outside the local namespace.
3042
This is done by checking if self.transport_server is a MemoryServer. if it
3043
is then we are chrooted already, if it is not then an HttpServer is used
3046
TODO RBC 20060127: make this an option to TestCaseWithTransport so it can
3047
be used without needed to redo it when a different
3048
subclass is in use ?
3052
from bzrlib.tests import http_server
3053
super(ChrootedTestCase, self).setUp()
3054
if not self.vfs_transport_factory == memory.MemoryServer:
3055
self.transport_readonly_server = http_server.HttpServer
3058
def condition_id_re(pattern):
3059
"""Create a condition filter which performs a re check on a test's id.
3061
:param pattern: A regular expression string.
3062
:return: A callable that returns True if the re matches.
3064
filter_re = re.compile(pattern, 0)
3065
def condition(test):
3067
return filter_re.search(test_id)
3071
def condition_isinstance(klass_or_klass_list):
3072
"""Create a condition filter which returns isinstance(param, klass).
3074
:return: A callable which when called with one parameter obj return the
3075
result of isinstance(obj, klass_or_klass_list).
3078
return isinstance(obj, klass_or_klass_list)
3082
def condition_id_in_list(id_list):
3083
"""Create a condition filter which verify that test's id in a list.
3085
:param id_list: A TestIdList object.
3086
:return: A callable that returns True if the test's id appears in the list.
3088
def condition(test):
3089
return id_list.includes(test.id())
3093
def condition_id_startswith(starts):
3094
"""Create a condition filter verifying that test's id starts with a string.
3096
:param starts: A list of string.
3097
:return: A callable that returns True if the test's id starts with one of
3100
def condition(test):
3101
for start in starts:
3102
if test.id().startswith(start):
3108
def exclude_tests_by_condition(suite, condition):
3109
"""Create a test suite which excludes some tests from suite.
3111
:param suite: The suite to get tests from.
3112
:param condition: A callable whose result evaluates True when called with a
3113
test case which should be excluded from the result.
3114
:return: A suite which contains the tests found in suite that fail
3118
for test in iter_suite_tests(suite):
3119
if not condition(test):
3121
return TestUtil.TestSuite(result)
3124
def filter_suite_by_condition(suite, condition):
3125
"""Create a test suite by filtering another one.
3127
:param suite: The source suite.
3128
:param condition: A callable whose result evaluates True when called with a
3129
test case which should be included in the result.
3130
:return: A suite which contains the tests found in suite that pass
3134
for test in iter_suite_tests(suite):
3137
return TestUtil.TestSuite(result)
3140
def filter_suite_by_re(suite, pattern):
3141
"""Create a test suite by filtering another one.
3143
:param suite: the source suite
3144
:param pattern: pattern that names must match
3145
:returns: the newly created suite
3147
condition = condition_id_re(pattern)
3148
result_suite = filter_suite_by_condition(suite, condition)
3152
def filter_suite_by_id_list(suite, test_id_list):
3153
"""Create a test suite by filtering another one.
3155
:param suite: The source suite.
3156
:param test_id_list: A list of the test ids to keep as strings.
3157
:returns: the newly created suite
3159
condition = condition_id_in_list(test_id_list)
3160
result_suite = filter_suite_by_condition(suite, condition)
3164
def filter_suite_by_id_startswith(suite, start):
3165
"""Create a test suite by filtering another one.
3167
:param suite: The source suite.
3168
:param start: A list of string the test id must start with one of.
3169
:returns: the newly created suite
3171
condition = condition_id_startswith(start)
3172
result_suite = filter_suite_by_condition(suite, condition)
3176
def exclude_tests_by_re(suite, pattern):
3177
"""Create a test suite which excludes some tests from suite.
3179
:param suite: The suite to get tests from.
3180
:param pattern: A regular expression string. Test ids that match this
3181
pattern will be excluded from the result.
3182
:return: A TestSuite that contains all the tests from suite without the
3183
tests that matched pattern. The order of tests is the same as it was in
3186
return exclude_tests_by_condition(suite, condition_id_re(pattern))
3189
def preserve_input(something):
3190
"""A helper for performing test suite transformation chains.
3192
:param something: Anything you want to preserve.
3198
def randomize_suite(suite):
3199
"""Return a new TestSuite with suite's tests in random order.
3201
The tests in the input suite are flattened into a single suite in order to
3202
accomplish this. Any nested TestSuites are removed to provide global
3205
tests = list(iter_suite_tests(suite))
3206
random.shuffle(tests)
3207
return TestUtil.TestSuite(tests)
3210
def split_suite_by_condition(suite, condition):
3211
"""Split a test suite into two by a condition.
3213
:param suite: The suite to split.
3214
:param condition: The condition to match on. Tests that match this
3215
condition are returned in the first test suite, ones that do not match
3216
are in the second suite.
3217
:return: A tuple of two test suites, where the first contains tests from
3218
suite matching the condition, and the second contains the remainder
3219
from suite. The order within each output suite is the same as it was in
3224
for test in iter_suite_tests(suite):
3226
matched.append(test)
3228
did_not_match.append(test)
3229
return TestUtil.TestSuite(matched), TestUtil.TestSuite(did_not_match)
3232
def split_suite_by_re(suite, pattern):
3233
"""Split a test suite into two by a regular expression.
3235
:param suite: The suite to split.
3236
:param pattern: A regular expression string. Test ids that match this
3237
pattern will be in the first test suite returned, and the others in the
3238
second test suite returned.
3239
:return: A tuple of two test suites, where the first contains tests from
3240
suite matching pattern, and the second contains the remainder from
3241
suite. The order within each output suite is the same as it was in
3244
return split_suite_by_condition(suite, condition_id_re(pattern))
3247
def run_suite(suite, name='test', verbose=False, pattern=".*",
3248
stop_on_failure=False,
3249
transport=None, lsprof_timed=None, bench_history=None,
3250
matching_tests_first=None,
3253
exclude_pattern=None,
3256
suite_decorators=None,
3258
result_decorators=None,
3260
"""Run a test suite for bzr selftest.
3262
:param runner_class: The class of runner to use. Must support the
3263
constructor arguments passed by run_suite which are more than standard
3265
:return: A boolean indicating success.
3267
TestCase._gather_lsprof_in_benchmarks = lsprof_timed
3272
if runner_class is None:
3273
runner_class = TextTestRunner
3276
runner = runner_class(stream=stream,
3278
verbosity=verbosity,
3279
bench_history=bench_history,
3281
result_decorators=result_decorators,
3283
runner.stop_on_failure=stop_on_failure
3284
if isinstance(suite, unittest.TestSuite):
3285
# Empty out _tests list of passed suite and populate new TestSuite
3286
suite._tests[:], suite = [], TestSuite(suite)
3287
# built in decorator factories:
3289
random_order(random_seed, runner),
3290
exclude_tests(exclude_pattern),
3292
if matching_tests_first:
3293
decorators.append(tests_first(pattern))
3295
decorators.append(filter_tests(pattern))
3296
if suite_decorators:
3297
decorators.extend(suite_decorators)
3298
# tell the result object how many tests will be running: (except if
3299
# --parallel=fork is being used. Robert said he will provide a better
3300
# progress design later -- vila 20090817)
3301
if fork_decorator not in decorators:
3302
decorators.append(CountingDecorator)
3303
for decorator in decorators:
3304
suite = decorator(suite)
3306
# Done after test suite decoration to allow randomisation etc
3307
# to take effect, though that is of marginal benefit.
3309
stream.write("Listing tests only ...\n")
3310
for t in iter_suite_tests(suite):
3311
stream.write("%s\n" % (t.id()))
3313
result = runner.run(suite)
3315
return result.wasStrictlySuccessful()
3317
return result.wasSuccessful()
3320
# A registry where get() returns a suite decorator.
3321
parallel_registry = registry.Registry()
3324
def fork_decorator(suite):
3325
if getattr(os, "fork", None) is None:
3326
raise errors.BzrCommandError("platform does not support fork,"
3327
" try --parallel=subprocess instead.")
3328
concurrency = osutils.local_concurrency()
3329
if concurrency == 1:
3331
from testtools import ConcurrentTestSuite
3332
return ConcurrentTestSuite(suite, fork_for_tests)
3333
parallel_registry.register('fork', fork_decorator)
3336
def subprocess_decorator(suite):
3337
concurrency = osutils.local_concurrency()
3338
if concurrency == 1:
3340
from testtools import ConcurrentTestSuite
3341
return ConcurrentTestSuite(suite, reinvoke_for_tests)
3342
parallel_registry.register('subprocess', subprocess_decorator)
3345
def exclude_tests(exclude_pattern):
3346
"""Return a test suite decorator that excludes tests."""
3347
if exclude_pattern is None:
3348
return identity_decorator
3349
def decorator(suite):
3350
return ExcludeDecorator(suite, exclude_pattern)
3354
def filter_tests(pattern):
3356
return identity_decorator
3357
def decorator(suite):
3358
return FilterTestsDecorator(suite, pattern)
3362
def random_order(random_seed, runner):
3363
"""Return a test suite decorator factory for randomising tests order.
3365
:param random_seed: now, a string which casts to a long, or a long.
3366
:param runner: A test runner with a stream attribute to report on.
3368
if random_seed is None:
3369
return identity_decorator
3370
def decorator(suite):
3371
return RandomDecorator(suite, random_seed, runner.stream)
3375
def tests_first(pattern):
3377
return identity_decorator
3378
def decorator(suite):
3379
return TestFirstDecorator(suite, pattern)
3383
def identity_decorator(suite):
3388
class TestDecorator(TestUtil.TestSuite):
3389
"""A decorator for TestCase/TestSuite objects.
3391
Contains rather than flattening suite passed on construction
3394
def __init__(self, suite=None):
3395
super(TestDecorator, self).__init__()
3396
if suite is not None:
3399
# Don't need subclass run method with suite emptying
3400
run = unittest.TestSuite.run
3403
class CountingDecorator(TestDecorator):
3404
"""A decorator which calls result.progress(self.countTestCases)."""
3406
def run(self, result):
3407
progress_method = getattr(result, 'progress', None)
3408
if callable(progress_method):
3409
progress_method(self.countTestCases(), SUBUNIT_SEEK_SET)
3410
return super(CountingDecorator, self).run(result)
3413
class ExcludeDecorator(TestDecorator):
3414
"""A decorator which excludes test matching an exclude pattern."""
3416
def __init__(self, suite, exclude_pattern):
3417
super(ExcludeDecorator, self).__init__(
3418
exclude_tests_by_re(suite, exclude_pattern))
3421
class FilterTestsDecorator(TestDecorator):
3422
"""A decorator which filters tests to those matching a pattern."""
3424
def __init__(self, suite, pattern):
3425
super(FilterTestsDecorator, self).__init__(
3426
filter_suite_by_re(suite, pattern))
3429
class RandomDecorator(TestDecorator):
3430
"""A decorator which randomises the order of its tests."""
3432
def __init__(self, suite, random_seed, stream):
3433
random_seed = self.actual_seed(random_seed)
3434
stream.write("Randomizing test order using seed %s\n\n" %
3436
# Initialise the random number generator.
3437
random.seed(random_seed)
3438
super(RandomDecorator, self).__init__(randomize_suite(suite))
3441
def actual_seed(seed):
3443
# We convert the seed to a long to make it reuseable across
3444
# invocations (because the user can reenter it).
3445
return long(time.time())
3447
# Convert the seed to a long if we can
3450
except (TypeError, ValueError):
3455
class TestFirstDecorator(TestDecorator):
3456
"""A decorator which moves named tests to the front."""
3458
def __init__(self, suite, pattern):
3459
super(TestFirstDecorator, self).__init__()
3460
self.addTests(split_suite_by_re(suite, pattern))
3463
def partition_tests(suite, count):
3464
"""Partition suite into count lists of tests."""
3465
# This just assigns tests in a round-robin fashion. On one hand this
3466
# splits up blocks of related tests that might run faster if they shared
3467
# resources, but on the other it avoids assigning blocks of slow tests to
3468
# just one partition. So the slowest partition shouldn't be much slower
3470
partitions = [list() for i in range(count)]
3471
tests = iter_suite_tests(suite)
3472
for partition, test in itertools.izip(itertools.cycle(partitions), tests):
3473
partition.append(test)
3477
def workaround_zealous_crypto_random():
3478
"""Crypto.Random want to help us being secure, but we don't care here.
3480
This workaround some test failure related to the sftp server. Once paramiko
3481
stop using the controversial API in Crypto.Random, we may get rid of it.
3484
from Crypto.Random import atfork
3490
def fork_for_tests(suite):
3491
"""Take suite and start up one runner per CPU by forking()
3493
:return: An iterable of TestCase-like objects which can each have
3494
run(result) called on them to feed tests to result.
3496
concurrency = osutils.local_concurrency()
3498
from subunit import ProtocolTestCase
3499
from subunit.test_results import AutoTimingTestResultDecorator
3500
class TestInOtherProcess(ProtocolTestCase):
3501
# Should be in subunit, I think. RBC.
3502
def __init__(self, stream, pid):
3503
ProtocolTestCase.__init__(self, stream)
3506
def run(self, result):
3508
ProtocolTestCase.run(self, result)
3510
pid, status = os.waitpid(self.pid, 0)
3511
# GZ 2011-10-18: If status is nonzero, should report to the result
3512
# that something went wrong.
3514
test_blocks = partition_tests(suite, concurrency)
3515
# Clear the tests from the original suite so it doesn't keep them alive
3516
suite._tests[:] = []
3517
for process_tests in test_blocks:
3518
process_suite = TestUtil.TestSuite(process_tests)
3519
# Also clear each split list so new suite has only reference
3520
process_tests[:] = []
3521
c2pread, c2pwrite = os.pipe()
3525
stream = os.fdopen(c2pwrite, 'wb', 1)
3526
workaround_zealous_crypto_random()
3528
# Leave stderr and stdout open so we can see test noise
3529
# Close stdin so that the child goes away if it decides to
3530
# read from stdin (otherwise its a roulette to see what
3531
# child actually gets keystrokes for pdb etc).
3533
subunit_result = AutoTimingTestResultDecorator(
3534
SubUnitBzrProtocolClient(stream))
3535
process_suite.run(subunit_result)
3537
# Try and report traceback on stream, but exit with error even
3538
# if stream couldn't be created or something else goes wrong.
3539
# The traceback is formatted to a string and written in one go
3540
# to avoid interleaving lines from multiple failing children.
3542
stream.write(traceback.format_exc())
3548
stream = os.fdopen(c2pread, 'rb', 1)
3549
test = TestInOtherProcess(stream, pid)
3554
def reinvoke_for_tests(suite):
3555
"""Take suite and start up one runner per CPU using subprocess().
3557
:return: An iterable of TestCase-like objects which can each have
3558
run(result) called on them to feed tests to result.
3560
concurrency = osutils.local_concurrency()
3562
from subunit import ProtocolTestCase
3563
class TestInSubprocess(ProtocolTestCase):
3564
def __init__(self, process, name):
3565
ProtocolTestCase.__init__(self, process.stdout)
3566
self.process = process
3567
self.process.stdin.close()
3570
def run(self, result):
3572
ProtocolTestCase.run(self, result)
3575
os.unlink(self.name)
3576
# print "pid %d finished" % finished_process
3577
test_blocks = partition_tests(suite, concurrency)
3578
for process_tests in test_blocks:
3579
# ugly; currently reimplement rather than reuses TestCase methods.
3580
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
3581
if not os.path.isfile(bzr_path):
3582
# We are probably installed. Assume sys.argv is the right file
3583
bzr_path = sys.argv[0]
3584
bzr_path = [bzr_path]
3585
if sys.platform == "win32":
3586
# if we're on windows, we can't execute the bzr script directly
3587
bzr_path = [sys.executable] + bzr_path
3588
fd, test_list_file_name = tempfile.mkstemp()
3589
test_list_file = os.fdopen(fd, 'wb', 1)
3590
for test in process_tests:
3591
test_list_file.write(test.id() + '\n')
3592
test_list_file.close()
3594
argv = bzr_path + ['selftest', '--load-list', test_list_file_name,
3596
if '--no-plugins' in sys.argv:
3597
argv.append('--no-plugins')
3598
# stderr=subprocess.STDOUT would be ideal, but until we prevent
3599
# noise on stderr it can interrupt the subunit protocol.
3600
process = subprocess.Popen(argv, stdin=subprocess.PIPE,
3601
stdout=subprocess.PIPE,
3602
stderr=subprocess.PIPE,
3604
test = TestInSubprocess(process, test_list_file_name)
3607
os.unlink(test_list_file_name)
3612
class ProfileResult(testtools.ExtendedToOriginalDecorator):
3613
"""Generate profiling data for all activity between start and success.
3615
The profile data is appended to the test's _benchcalls attribute and can
3616
be accessed by the forwarded-to TestResult.
3618
While it might be cleaner do accumulate this in stopTest, addSuccess is
3619
where our existing output support for lsprof is, and this class aims to
3620
fit in with that: while it could be moved it's not necessary to accomplish
3621
test profiling, nor would it be dramatically cleaner.
3624
def startTest(self, test):
3625
self.profiler = bzrlib.lsprof.BzrProfiler()
3626
# Prevent deadlocks in tests that use lsprof: those tests will
3628
bzrlib.lsprof.BzrProfiler.profiler_block = 0
3629
self.profiler.start()
3630
testtools.ExtendedToOriginalDecorator.startTest(self, test)
3632
def addSuccess(self, test):
3633
stats = self.profiler.stop()
3635
calls = test._benchcalls
3636
except AttributeError:
3637
test._benchcalls = []
3638
calls = test._benchcalls
3639
calls.append(((test.id(), "", ""), stats))
3640
testtools.ExtendedToOriginalDecorator.addSuccess(self, test)
3642
def stopTest(self, test):
3643
testtools.ExtendedToOriginalDecorator.stopTest(self, test)
3644
self.profiler = None
3647
# Controlled by "bzr selftest -E=..." option
3648
# Currently supported:
3649
# -Eallow_debug Will no longer clear debug.debug_flags() so it
3650
# preserves any flags supplied at the command line.
3651
# -Edisable_lock_checks Turns errors in mismatched locks into simple prints
3652
# rather than failing tests. And no longer raise
3653
# LockContention when fctnl locks are not being used
3654
# with proper exclusion rules.
3655
# -Ethreads Will display thread ident at creation/join time to
3656
# help track thread leaks
3657
# -Euncollected_cases Display the identity of any test cases that weren't
3658
# deallocated after being completed.
3659
# -Econfig_stats Will collect statistics using addDetail
3660
selftest_debug_flags = set()
3663
def selftest(verbose=False, pattern=".*", stop_on_failure=True,
3665
test_suite_factory=None,
3668
matching_tests_first=None,
3671
exclude_pattern=None,
3677
suite_decorators=None,
3681
"""Run the whole test suite under the enhanced runner"""
3682
# XXX: Very ugly way to do this...
3683
# Disable warning about old formats because we don't want it to disturb
3684
# any blackbox tests.
3685
from bzrlib import repository
3686
repository._deprecation_warning_done = True
3688
global default_transport
3689
if transport is None:
3690
transport = default_transport
3691
old_transport = default_transport
3692
default_transport = transport
3693
global selftest_debug_flags
3694
old_debug_flags = selftest_debug_flags
3695
if debug_flags is not None:
3696
selftest_debug_flags = set(debug_flags)
3698
if load_list is None:
3701
keep_only = load_test_id_list(load_list)
3703
starting_with = [test_prefix_alias_registry.resolve_alias(start)
3704
for start in starting_with]
3705
if test_suite_factory is None:
3706
# Reduce loading time by loading modules based on the starting_with
3708
suite = test_suite(keep_only, starting_with)
3710
suite = test_suite_factory()
3712
# But always filter as requested.
3713
suite = filter_suite_by_id_startswith(suite, starting_with)
3714
result_decorators = []
3716
result_decorators.append(ProfileResult)
3717
return run_suite(suite, 'testbzr', verbose=verbose, pattern=pattern,
3718
stop_on_failure=stop_on_failure,
3719
transport=transport,
3720
lsprof_timed=lsprof_timed,
3721
bench_history=bench_history,
3722
matching_tests_first=matching_tests_first,
3723
list_only=list_only,
3724
random_seed=random_seed,
3725
exclude_pattern=exclude_pattern,
3727
runner_class=runner_class,
3728
suite_decorators=suite_decorators,
3730
result_decorators=result_decorators,
3733
default_transport = old_transport
3734
selftest_debug_flags = old_debug_flags
3737
def load_test_id_list(file_name):
3738
"""Load a test id list from a text file.
3740
The format is one test id by line. No special care is taken to impose
3741
strict rules, these test ids are used to filter the test suite so a test id
3742
that do not match an existing test will do no harm. This allows user to add
3743
comments, leave blank lines, etc.
3747
ftest = open(file_name, 'rt')
3749
if e.errno != errno.ENOENT:
3752
raise errors.NoSuchFile(file_name)
3754
for test_name in ftest.readlines():
3755
test_list.append(test_name.strip())
3760
def suite_matches_id_list(test_suite, id_list):
3761
"""Warns about tests not appearing or appearing more than once.
3763
:param test_suite: A TestSuite object.
3764
:param test_id_list: The list of test ids that should be found in
3767
:return: (absents, duplicates) absents is a list containing the test found
3768
in id_list but not in test_suite, duplicates is a list containing the
3769
test found multiple times in test_suite.
3771
When using a prefined test id list, it may occurs that some tests do not
3772
exist anymore or that some tests use the same id. This function warns the
3773
tester about potential problems in his workflow (test lists are volatile)
3774
or in the test suite itself (using the same id for several tests does not
3775
help to localize defects).
3777
# Build a dict counting id occurrences
3779
for test in iter_suite_tests(test_suite):
3781
tests[id] = tests.get(id, 0) + 1
3786
occurs = tests.get(id, 0)
3788
not_found.append(id)
3790
duplicates.append(id)
3792
return not_found, duplicates
3795
class TestIdList(object):
3796
"""Test id list to filter a test suite.
3798
Relying on the assumption that test ids are built as:
3799
<module>[.<class>.<method>][(<param>+)], <module> being in python dotted
3800
notation, this class offers methods to :
3801
- avoid building a test suite for modules not refered to in the test list,
3802
- keep only the tests listed from the module test suite.
3805
def __init__(self, test_id_list):
3806
# When a test suite needs to be filtered against us we compare test ids
3807
# for equality, so a simple dict offers a quick and simple solution.
3808
self.tests = dict().fromkeys(test_id_list, True)
3810
# While unittest.TestCase have ids like:
3811
# <module>.<class>.<method>[(<param+)],
3812
# doctest.DocTestCase can have ids like:
3815
# <module>.<function>
3816
# <module>.<class>.<method>
3818
# Since we can't predict a test class from its name only, we settle on
3819
# a simple constraint: a test id always begins with its module name.
3822
for test_id in test_id_list:
3823
parts = test_id.split('.')
3824
mod_name = parts.pop(0)
3825
modules[mod_name] = True
3827
mod_name += '.' + part
3828
modules[mod_name] = True
3829
self.modules = modules
3831
def refers_to(self, module_name):
3832
"""Is there tests for the module or one of its sub modules."""
3833
return self.modules.has_key(module_name)
3835
def includes(self, test_id):
3836
return self.tests.has_key(test_id)
3839
class TestPrefixAliasRegistry(registry.Registry):
3840
"""A registry for test prefix aliases.
3842
This helps implement shorcuts for the --starting-with selftest
3843
option. Overriding existing prefixes is not allowed but not fatal (a
3844
warning will be emitted).
3847
def register(self, key, obj, help=None, info=None,
3848
override_existing=False):
3849
"""See Registry.register.
3851
Trying to override an existing alias causes a warning to be emitted,
3852
not a fatal execption.
3855
super(TestPrefixAliasRegistry, self).register(
3856
key, obj, help=help, info=info, override_existing=False)
3858
actual = self.get(key)
3860
'Test prefix alias %s is already used for %s, ignoring %s'
3861
% (key, actual, obj))
3863
def resolve_alias(self, id_start):
3864
"""Replace the alias by the prefix in the given string.
3866
Using an unknown prefix is an error to help catching typos.
3868
parts = id_start.split('.')
3870
parts[0] = self.get(parts[0])
3872
raise errors.BzrCommandError(
3873
'%s is not a known test prefix alias' % parts[0])
3874
return '.'.join(parts)
3877
test_prefix_alias_registry = TestPrefixAliasRegistry()
3878
"""Registry of test prefix aliases."""
3881
# This alias allows to detect typos ('bzrlin.') by making all valid test ids
3882
# appear prefixed ('bzrlib.' is "replaced" by 'bzrlib.').
3883
test_prefix_alias_registry.register('bzrlib', 'bzrlib')
3885
# Obvious highest levels prefixes, feel free to add your own via a plugin
3886
test_prefix_alias_registry.register('bd', 'bzrlib.doc')
3887
test_prefix_alias_registry.register('bu', 'bzrlib.utils')
3888
test_prefix_alias_registry.register('bt', 'bzrlib.tests')
3889
test_prefix_alias_registry.register('bb', 'bzrlib.tests.blackbox')
3890
test_prefix_alias_registry.register('bp', 'bzrlib.plugins')
3893
def _test_suite_testmod_names():
3894
"""Return the standard list of test module names to test."""
3897
'bzrlib.tests.blackbox',
3898
'bzrlib.tests.commands',
3899
'bzrlib.tests.doc_generate',
3900
'bzrlib.tests.per_branch',
3901
'bzrlib.tests.per_bzrdir',
3902
'bzrlib.tests.per_controldir',
3903
'bzrlib.tests.per_controldir_colo',
3904
'bzrlib.tests.per_foreign_vcs',
3905
'bzrlib.tests.per_interrepository',
3906
'bzrlib.tests.per_intertree',
3907
'bzrlib.tests.per_inventory',
3908
'bzrlib.tests.per_interbranch',
3909
'bzrlib.tests.per_lock',
3910
'bzrlib.tests.per_merger',
3911
'bzrlib.tests.per_transport',
3912
'bzrlib.tests.per_tree',
3913
'bzrlib.tests.per_pack_repository',
3914
'bzrlib.tests.per_repository',
3915
'bzrlib.tests.per_repository_chk',
3916
'bzrlib.tests.per_repository_reference',
3917
'bzrlib.tests.per_repository_vf',
3918
'bzrlib.tests.per_uifactory',
3919
'bzrlib.tests.per_versionedfile',
3920
'bzrlib.tests.per_workingtree',
3921
'bzrlib.tests.test__annotator',
3922
'bzrlib.tests.test__bencode',
3923
'bzrlib.tests.test__btree_serializer',
3924
'bzrlib.tests.test__chk_map',
3925
'bzrlib.tests.test__dirstate_helpers',
3926
'bzrlib.tests.test__groupcompress',
3927
'bzrlib.tests.test__known_graph',
3928
'bzrlib.tests.test__rio',
3929
'bzrlib.tests.test__simple_set',
3930
'bzrlib.tests.test__static_tuple',
3931
'bzrlib.tests.test__walkdirs_win32',
3932
'bzrlib.tests.test_ancestry',
3933
'bzrlib.tests.test_annotate',
3934
'bzrlib.tests.test_api',
3935
'bzrlib.tests.test_atomicfile',
3936
'bzrlib.tests.test_bad_files',
3937
'bzrlib.tests.test_bisect_multi',
3938
'bzrlib.tests.test_branch',
3939
'bzrlib.tests.test_branchbuilder',
3940
'bzrlib.tests.test_btree_index',
3941
'bzrlib.tests.test_bugtracker',
3942
'bzrlib.tests.test_bundle',
3943
'bzrlib.tests.test_bzrdir',
3944
'bzrlib.tests.test__chunks_to_lines',
3945
'bzrlib.tests.test_cache_utf8',
3946
'bzrlib.tests.test_chk_map',
3947
'bzrlib.tests.test_chk_serializer',
3948
'bzrlib.tests.test_chunk_writer',
3949
'bzrlib.tests.test_clean_tree',
3950
'bzrlib.tests.test_cleanup',
3951
'bzrlib.tests.test_cmdline',
3952
'bzrlib.tests.test_commands',
3953
'bzrlib.tests.test_commit',
3954
'bzrlib.tests.test_commit_merge',
3955
'bzrlib.tests.test_config',
3956
'bzrlib.tests.test_conflicts',
3957
'bzrlib.tests.test_controldir',
3958
'bzrlib.tests.test_counted_lock',
3959
'bzrlib.tests.test_crash',
3960
'bzrlib.tests.test_decorators',
3961
'bzrlib.tests.test_delta',
3962
'bzrlib.tests.test_debug',
3963
'bzrlib.tests.test_diff',
3964
'bzrlib.tests.test_directory_service',
3965
'bzrlib.tests.test_dirstate',
3966
'bzrlib.tests.test_email_message',
3967
'bzrlib.tests.test_eol_filters',
3968
'bzrlib.tests.test_errors',
3969
'bzrlib.tests.test_estimate_compressed_size',
3970
'bzrlib.tests.test_export',
3971
'bzrlib.tests.test_export_pot',
3972
'bzrlib.tests.test_extract',
3973
'bzrlib.tests.test_features',
3974
'bzrlib.tests.test_fetch',
3975
'bzrlib.tests.test_fixtures',
3976
'bzrlib.tests.test_fifo_cache',
3977
'bzrlib.tests.test_filters',
3978
'bzrlib.tests.test_filter_tree',
3979
'bzrlib.tests.test_ftp_transport',
3980
'bzrlib.tests.test_foreign',
3981
'bzrlib.tests.test_generate_docs',
3982
'bzrlib.tests.test_generate_ids',
3983
'bzrlib.tests.test_globbing',
3984
'bzrlib.tests.test_gpg',
3985
'bzrlib.tests.test_graph',
3986
'bzrlib.tests.test_groupcompress',
3987
'bzrlib.tests.test_hashcache',
3988
'bzrlib.tests.test_help',
3989
'bzrlib.tests.test_hooks',
3990
'bzrlib.tests.test_http',
3991
'bzrlib.tests.test_http_response',
3992
'bzrlib.tests.test_https_ca_bundle',
3993
'bzrlib.tests.test_i18n',
3994
'bzrlib.tests.test_identitymap',
3995
'bzrlib.tests.test_ignores',
3996
'bzrlib.tests.test_index',
3997
'bzrlib.tests.test_import_tariff',
3998
'bzrlib.tests.test_info',
3999
'bzrlib.tests.test_inv',
4000
'bzrlib.tests.test_inventory_delta',
4001
'bzrlib.tests.test_knit',
4002
'bzrlib.tests.test_lazy_import',
4003
'bzrlib.tests.test_lazy_regex',
4004
'bzrlib.tests.test_library_state',
4005
'bzrlib.tests.test_lock',
4006
'bzrlib.tests.test_lockable_files',
4007
'bzrlib.tests.test_lockdir',
4008
'bzrlib.tests.test_log',
4009
'bzrlib.tests.test_lru_cache',
4010
'bzrlib.tests.test_lsprof',
4011
'bzrlib.tests.test_mail_client',
4012
'bzrlib.tests.test_matchers',
4013
'bzrlib.tests.test_memorytree',
4014
'bzrlib.tests.test_merge',
4015
'bzrlib.tests.test_merge3',
4016
'bzrlib.tests.test_merge_core',
4017
'bzrlib.tests.test_merge_directive',
4018
'bzrlib.tests.test_mergetools',
4019
'bzrlib.tests.test_missing',
4020
'bzrlib.tests.test_msgeditor',
4021
'bzrlib.tests.test_multiparent',
4022
'bzrlib.tests.test_mutabletree',
4023
'bzrlib.tests.test_nonascii',
4024
'bzrlib.tests.test_options',
4025
'bzrlib.tests.test_osutils',
4026
'bzrlib.tests.test_osutils_encodings',
4027
'bzrlib.tests.test_pack',
4028
'bzrlib.tests.test_patch',
4029
'bzrlib.tests.test_patches',
4030
'bzrlib.tests.test_permissions',
4031
'bzrlib.tests.test_plugins',
4032
'bzrlib.tests.test_progress',
4033
'bzrlib.tests.test_pyutils',
4034
'bzrlib.tests.test_read_bundle',
4035
'bzrlib.tests.test_reconcile',
4036
'bzrlib.tests.test_reconfigure',
4037
'bzrlib.tests.test_registry',
4038
'bzrlib.tests.test_remote',
4039
'bzrlib.tests.test_rename_map',
4040
'bzrlib.tests.test_repository',
4041
'bzrlib.tests.test_revert',
4042
'bzrlib.tests.test_revision',
4043
'bzrlib.tests.test_revisionspec',
4044
'bzrlib.tests.test_revisiontree',
4045
'bzrlib.tests.test_rio',
4046
'bzrlib.tests.test_rules',
4047
'bzrlib.tests.test_sampler',
4048
'bzrlib.tests.test_scenarios',
4049
'bzrlib.tests.test_script',
4050
'bzrlib.tests.test_selftest',
4051
'bzrlib.tests.test_serializer',
4052
'bzrlib.tests.test_setup',
4053
'bzrlib.tests.test_sftp_transport',
4054
'bzrlib.tests.test_shelf',
4055
'bzrlib.tests.test_shelf_ui',
4056
'bzrlib.tests.test_smart',
4057
'bzrlib.tests.test_smart_add',
4058
'bzrlib.tests.test_smart_request',
4059
'bzrlib.tests.test_smart_signals',
4060
'bzrlib.tests.test_smart_transport',
4061
'bzrlib.tests.test_smtp_connection',
4062
'bzrlib.tests.test_source',
4063
'bzrlib.tests.test_ssh_transport',
4064
'bzrlib.tests.test_status',
4065
'bzrlib.tests.test_store',
4066
'bzrlib.tests.test_strace',
4067
'bzrlib.tests.test_subsume',
4068
'bzrlib.tests.test_switch',
4069
'bzrlib.tests.test_symbol_versioning',
4070
'bzrlib.tests.test_tag',
4071
'bzrlib.tests.test_test_server',
4072
'bzrlib.tests.test_testament',
4073
'bzrlib.tests.test_textfile',
4074
'bzrlib.tests.test_textmerge',
4075
'bzrlib.tests.test_cethread',
4076
'bzrlib.tests.test_timestamp',
4077
'bzrlib.tests.test_trace',
4078
'bzrlib.tests.test_transactions',
4079
'bzrlib.tests.test_transform',
4080
'bzrlib.tests.test_transport',
4081
'bzrlib.tests.test_transport_log',
4082
'bzrlib.tests.test_tree',
4083
'bzrlib.tests.test_treebuilder',
4084
'bzrlib.tests.test_treeshape',
4085
'bzrlib.tests.test_tsort',
4086
'bzrlib.tests.test_tuned_gzip',
4087
'bzrlib.tests.test_ui',
4088
'bzrlib.tests.test_uncommit',
4089
'bzrlib.tests.test_upgrade',
4090
'bzrlib.tests.test_upgrade_stacked',
4091
'bzrlib.tests.test_urlutils',
4092
'bzrlib.tests.test_utextwrap',
4093
'bzrlib.tests.test_version',
4094
'bzrlib.tests.test_version_info',
4095
'bzrlib.tests.test_versionedfile',
4096
'bzrlib.tests.test_vf_search',
4097
'bzrlib.tests.test_weave',
4098
'bzrlib.tests.test_whitebox',
4099
'bzrlib.tests.test_win32utils',
4100
'bzrlib.tests.test_workingtree',
4101
'bzrlib.tests.test_workingtree_4',
4102
'bzrlib.tests.test_wsgi',
4103
'bzrlib.tests.test_xml',
4107
def _test_suite_modules_to_doctest():
4108
"""Return the list of modules to doctest."""
4110
# GZ 2009-03-31: No docstrings with -OO so there's nothing to doctest
4114
'bzrlib.branchbuilder',
4115
'bzrlib.decorators',
4117
'bzrlib.iterablefile',
4122
'bzrlib.symbol_versioning',
4124
'bzrlib.tests.fixtures',
4126
'bzrlib.transport.http',
4127
'bzrlib.version_info_formats.format_custom',
4131
def test_suite(keep_only=None, starting_with=None):
4132
"""Build and return TestSuite for the whole of bzrlib.
4134
:param keep_only: A list of test ids limiting the suite returned.
4136
:param starting_with: An id limiting the suite returned to the tests
4139
This function can be replaced if you need to change the default test
4140
suite on a global basis, but it is not encouraged.
4143
loader = TestUtil.TestLoader()
4145
if keep_only is not None:
4146
id_filter = TestIdList(keep_only)
4148
# We take precedence over keep_only because *at loading time* using
4149
# both options means we will load less tests for the same final result.
4150
def interesting_module(name):
4151
for start in starting_with:
4153
# Either the module name starts with the specified string
4154
name.startswith(start)
4155
# or it may contain tests starting with the specified string
4156
or start.startswith(name)
4160
loader = TestUtil.FilteredByModuleTestLoader(interesting_module)
4162
elif keep_only is not None:
4163
loader = TestUtil.FilteredByModuleTestLoader(id_filter.refers_to)
4164
def interesting_module(name):
4165
return id_filter.refers_to(name)
4168
loader = TestUtil.TestLoader()
4169
def interesting_module(name):
4170
# No filtering, all modules are interesting
4173
suite = loader.suiteClass()
4175
# modules building their suite with loadTestsFromModuleNames
4176
suite.addTest(loader.loadTestsFromModuleNames(_test_suite_testmod_names()))
4178
for mod in _test_suite_modules_to_doctest():
4179
if not interesting_module(mod):
4180
# No tests to keep here, move along
4183
# note that this really does mean "report only" -- doctest
4184
# still runs the rest of the examples
4185
doc_suite = IsolatedDocTestSuite(
4186
mod, optionflags=doctest.REPORT_ONLY_FIRST_FAILURE)
4187
except ValueError, e:
4188
print '**failed to get doctest for: %s\n%s' % (mod, e)
4190
if len(doc_suite._tests) == 0:
4191
raise errors.BzrError("no doctests found in %s" % (mod,))
4192
suite.addTest(doc_suite)
4194
default_encoding = sys.getdefaultencoding()
4195
for name, plugin in _mod_plugin.plugins().items():
4196
if not interesting_module(plugin.module.__name__):
4198
plugin_suite = plugin.test_suite()
4199
# We used to catch ImportError here and turn it into just a warning,
4200
# but really if you don't have --no-plugins this should be a failure.
4201
# mbp 20080213 - see http://bugs.launchpad.net/bugs/189771
4202
if plugin_suite is None:
4203
plugin_suite = plugin.load_plugin_tests(loader)
4204
if plugin_suite is not None:
4205
suite.addTest(plugin_suite)
4206
if default_encoding != sys.getdefaultencoding():
4208
'Plugin "%s" tried to reset default encoding to: %s', name,
4209
sys.getdefaultencoding())
4211
sys.setdefaultencoding(default_encoding)
4213
if keep_only is not None:
4214
# Now that the referred modules have loaded their tests, keep only the
4216
suite = filter_suite_by_id_list(suite, id_filter)
4217
# Do some sanity checks on the id_list filtering
4218
not_found, duplicates = suite_matches_id_list(suite, keep_only)
4220
# The tester has used both keep_only and starting_with, so he is
4221
# already aware that some tests are excluded from the list, there
4222
# is no need to tell him which.
4225
# Some tests mentioned in the list are not in the test suite. The
4226
# list may be out of date, report to the tester.
4227
for id in not_found:
4228
trace.warning('"%s" not found in the test suite', id)
4229
for id in duplicates:
4230
trace.warning('"%s" is used as an id by several tests', id)
4235
def multiply_scenarios(*scenarios):
4236
"""Multiply two or more iterables of scenarios.
4238
It is safe to pass scenario generators or iterators.
4240
:returns: A list of compound scenarios: the cross-product of all
4241
scenarios, with the names concatenated and the parameters
4244
return reduce(_multiply_two_scenarios, map(list, scenarios))
4247
def _multiply_two_scenarios(scenarios_left, scenarios_right):
4248
"""Multiply two sets of scenarios.
4250
:returns: the cartesian product of the two sets of scenarios, that is
4251
a scenario for every possible combination of a left scenario and a
4255
('%s,%s' % (left_name, right_name),
4256
dict(left_dict.items() + right_dict.items()))
4257
for left_name, left_dict in scenarios_left
4258
for right_name, right_dict in scenarios_right]
4261
def multiply_tests(tests, scenarios, result):
4262
"""Multiply tests_list by scenarios into result.
4264
This is the core workhorse for test parameterisation.
4266
Typically the load_tests() method for a per-implementation test suite will
4267
call multiply_tests and return the result.
4269
:param tests: The tests to parameterise.
4270
:param scenarios: The scenarios to apply: pairs of (scenario_name,
4271
scenario_param_dict).
4272
:param result: A TestSuite to add created tests to.
4274
This returns the passed in result TestSuite with the cross product of all
4275
the tests repeated once for each scenario. Each test is adapted by adding
4276
the scenario name at the end of its id(), and updating the test object's
4277
__dict__ with the scenario_param_dict.
4279
>>> import bzrlib.tests.test_sampler
4280
>>> r = multiply_tests(
4281
... bzrlib.tests.test_sampler.DemoTest('test_nothing'),
4282
... [('one', dict(param=1)),
4283
... ('two', dict(param=2))],
4284
... TestUtil.TestSuite())
4285
>>> tests = list(iter_suite_tests(r))
4289
'bzrlib.tests.test_sampler.DemoTest.test_nothing(one)'
4295
for test in iter_suite_tests(tests):
4296
apply_scenarios(test, scenarios, result)
4300
def apply_scenarios(test, scenarios, result):
4301
"""Apply the scenarios in scenarios to test and add to result.
4303
:param test: The test to apply scenarios to.
4304
:param scenarios: An iterable of scenarios to apply to test.
4306
:seealso: apply_scenario
4308
for scenario in scenarios:
4309
result.addTest(apply_scenario(test, scenario))
4313
def apply_scenario(test, scenario):
4314
"""Copy test and apply scenario to it.
4316
:param test: A test to adapt.
4317
:param scenario: A tuple describing the scenarion.
4318
The first element of the tuple is the new test id.
4319
The second element is a dict containing attributes to set on the
4321
:return: The adapted test.
4323
new_id = "%s(%s)" % (test.id(), scenario[0])
4324
new_test = clone_test(test, new_id)
4325
for name, value in scenario[1].items():
4326
setattr(new_test, name, value)
4330
def clone_test(test, new_id):
4331
"""Clone a test giving it a new id.
4333
:param test: The test to clone.
4334
:param new_id: The id to assign to it.
4335
:return: The new test.
4337
new_test = copy.copy(test)
4338
new_test.id = lambda: new_id
4339
# XXX: Workaround <https://bugs.launchpad.net/testtools/+bug/637725>, which
4340
# causes cloned tests to share the 'details' dict. This makes it hard to
4341
# read the test output for parameterized tests, because tracebacks will be
4342
# associated with irrelevant tests.
4344
details = new_test._TestCase__details
4345
except AttributeError:
4346
# must be a different version of testtools than expected. Do nothing.
4349
# Reset the '__details' dict.
4350
new_test._TestCase__details = {}
4354
def permute_tests_for_extension(standard_tests, loader, py_module_name,
4356
"""Helper for permutating tests against an extension module.
4358
This is meant to be used inside a modules 'load_tests()' function. It will
4359
create 2 scenarios, and cause all tests in the 'standard_tests' to be run
4360
against both implementations. Setting 'test.module' to the appropriate
4361
module. See bzrlib.tests.test__chk_map.load_tests as an example.
4363
:param standard_tests: A test suite to permute
4364
:param loader: A TestLoader
4365
:param py_module_name: The python path to a python module that can always
4366
be loaded, and will be considered the 'python' implementation. (eg
4367
'bzrlib._chk_map_py')
4368
:param ext_module_name: The python path to an extension module. If the
4369
module cannot be loaded, a single test will be added, which notes that
4370
the module is not available. If it can be loaded, all standard_tests
4371
will be run against that module.
4372
:return: (suite, feature) suite is a test-suite that has all the permuted
4373
tests. feature is the Feature object that can be used to determine if
4374
the module is available.
4377
from bzrlib.tests.features import ModuleAvailableFeature
4378
py_module = pyutils.get_named_object(py_module_name)
4380
('python', {'module': py_module}),
4382
suite = loader.suiteClass()
4383
feature = ModuleAvailableFeature(ext_module_name)
4384
if feature.available():
4385
scenarios.append(('C', {'module': feature.module}))
4387
# the compiled module isn't available, so we add a failing test
4388
class FailWithoutFeature(TestCase):
4389
def test_fail(self):
4390
self.requireFeature(feature)
4391
suite.addTest(loader.loadTestsFromTestCase(FailWithoutFeature))
4392
result = multiply_tests(standard_tests, scenarios, suite)
4393
return result, feature
4396
def _rmtree_temp_dir(dirname, test_id=None):
4397
# If LANG=C we probably have created some bogus paths
4398
# which rmtree(unicode) will fail to delete
4399
# so make sure we are using rmtree(str) to delete everything
4400
# except on win32, where rmtree(str) will fail
4401
# since it doesn't have the property of byte-stream paths
4402
# (they are either ascii or mbcs)
4403
if sys.platform == 'win32':
4404
# make sure we are using the unicode win32 api
4405
dirname = unicode(dirname)
4407
dirname = dirname.encode(sys.getfilesystemencoding())
4409
osutils.rmtree(dirname)
4411
# We don't want to fail here because some useful display will be lost
4412
# otherwise. Polluting the tmp dir is bad, but not giving all the
4413
# possible info to the test runner is even worse.
4415
ui.ui_factory.clear_term()
4416
sys.stderr.write('\nWhile running: %s\n' % (test_id,))
4417
# Ugly, but the last thing we want here is fail, so bear with it.
4418
printable_e = str(e).decode(osutils.get_user_encoding(), 'replace'
4419
).encode('ascii', 'replace')
4420
sys.stderr.write('Unable to remove testing dir %s\n%s'
4421
% (os.path.basename(dirname), printable_e))
4424
def probe_unicode_in_user_encoding():
4425
"""Try to encode several unicode strings to use in unicode-aware tests.
4426
Return first successfull match.
4428
:return: (unicode value, encoded plain string value) or (None, None)
4430
possible_vals = [u'm\xb5', u'\xe1', u'\u0410']
4431
for uni_val in possible_vals:
4433
str_val = uni_val.encode(osutils.get_user_encoding())
4434
except UnicodeEncodeError:
4435
# Try a different character
4438
return uni_val, str_val
4442
def probe_bad_non_ascii(encoding):
4443
"""Try to find [bad] character with code [128..255]
4444
that cannot be decoded to unicode in some encoding.
4445
Return None if all non-ascii characters is valid
4448
for i in xrange(128, 256):
4451
char.decode(encoding)
4452
except UnicodeDecodeError:
4457
# Only define SubUnitBzrRunner if subunit is available.
4459
from subunit import TestProtocolClient
4460
from subunit.test_results import AutoTimingTestResultDecorator
4461
class SubUnitBzrProtocolClient(TestProtocolClient):
4463
def stopTest(self, test):
4464
super(SubUnitBzrProtocolClient, self).stopTest(test)
4465
_clear__type_equality_funcs(test)
4467
def addSuccess(self, test, details=None):
4468
# The subunit client always includes the details in the subunit
4469
# stream, but we don't want to include it in ours.
4470
if details is not None and 'log' in details:
4472
return super(SubUnitBzrProtocolClient, self).addSuccess(
4475
class SubUnitBzrRunner(TextTestRunner):
4476
def run(self, test):
4477
result = AutoTimingTestResultDecorator(
4478
SubUnitBzrProtocolClient(self.stream))
4485
# API compatibility for old plugins; see bug 892622.
4488
'HTTPServerFeature',
4489
'ModuleAvailableFeature',
4490
'HTTPSServerFeature', 'SymlinkFeature', 'HardlinkFeature',
4491
'OsFifoFeature', 'UnicodeFilenameFeature',
4492
'ByteStringNamedFilesystem', 'UTF8Filesystem',
4493
'BreakinFeature', 'CaseInsCasePresFilenameFeature',
4494
'CaseInsensitiveFilesystemFeature', 'case_sensitive_filesystem_feature',
4495
'posix_permissions_feature',
4497
globals()[name] = _CompatabilityThunkFeature(
4498
symbol_versioning.deprecated_in((2, 5, 0)),
4499
'bzrlib.tests', name,
4500
name, 'bzrlib.tests.features')
4503
for (old_name, new_name) in [
4504
('UnicodeFilename', 'UnicodeFilenameFeature'),
4506
globals()[name] = _CompatabilityThunkFeature(
4507
symbol_versioning.deprecated_in((2, 5, 0)),
4508
'bzrlib.tests', old_name,
4509
new_name, 'bzrlib.tests.features')
added
removed
bzrlib/tests/__init__.py
