1
# Copyright (C) 2005 by Canonical Ltd
1
# Copyright (C) 2005, 2006, 2007, 2008, 2009 Canonical Ltd
3
3
# This program is free software; you can redistribute it and/or modify
4
4
# it under the terms of the GNU General Public License as published by
5
5
# the Free Software Foundation; either version 2 of the License, or
6
6
# (at your option) any later version.
8
8
# This program is distributed in the hope that it will be useful,
9
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
11
# GNU General Public License for more details.
13
13
# You should have received a copy of the GNU General Public License
14
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18
from unittest import TestResult, TestCase
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
# TODO: Perhaps there should be an API to find out if bzr running under the
19
# test suite -- some plugins might want to avoid making intrusive changes if
20
# this is the case. However, we want behaviour under to test to diverge as
21
# little as possible, so this should be used rarely if it's added at all.
22
# (Suggestion from j-a-meinel, 2005-11-24)
24
# NOTE: Some classes in here use camelCaseNaming() rather than
25
# underscore_naming(). That's for consistency with unittest; it's not the
26
# general style of bzrlib. Please continue that consistency when adding e.g.
27
# new assertFoo() methods.
31
from cStringIO import StringIO
38
from pprint import pformat
43
from subprocess import Popen, PIPE, STDOUT
69
import bzrlib.commands
70
import bzrlib.timestamp
72
import bzrlib.inventory
73
import bzrlib.iterablefile
22
from subprocess import call, Popen, PIPE
23
except ImportError, e:
24
sys.stderr.write("testbzr: sorry, this test suite requires the subprocess module\n"
25
"this is shipped with python2.4 and available separately for 2.3\n")
78
# lsprof not available
80
from bzrlib.merge import merge_inner
83
from bzrlib.smart import client, request, server
85
from bzrlib import symbol_versioning
86
from bzrlib.symbol_versioning import (
93
from bzrlib.transport import get_transport
94
import bzrlib.transport
95
from bzrlib.transport.local import LocalURLServer
96
from bzrlib.transport.memory import MemoryServer
97
from bzrlib.transport.readonly import ReadonlyServer
98
from bzrlib.trace import mutter, note
99
from bzrlib.tests import TestUtil
100
from bzrlib.tests.http_server import HttpServer
101
from bzrlib.tests.TestUtil import (
105
from bzrlib.tests.treeshape import build_tree_contents
106
from bzrlib.ui import NullProgressView
107
from bzrlib.ui.text import TextUIFactory
108
import bzrlib.version_info_formats.format_custom
109
from bzrlib.workingtree import WorkingTree, WorkingTreeFormat2
111
# Mark this python module as being part of the implementation
112
# of unittest: this gives us better tracebacks where the last
113
# shown frame is the test code, not our assertXYZ.
116
default_transport = LocalURLServer
118
# Subunit result codes, defined here to prevent a hard dependency on subunit.
123
class ExtendedTestResult(unittest._TextTestResult):
124
"""Accepts, reports and accumulates the results of running tests.
126
Compared to the unittest version this class adds support for
127
profiling, benchmarking, stopping as soon as a test fails, and
128
skipping tests. There are further-specialized subclasses for
129
different types of display.
131
When a test finishes, in whatever way, it calls one of the addSuccess,
132
addFailure or addError classes. These in turn may redirect to a more
133
specific case for the special test results supported by our extended
136
Note that just one of these objects is fed the results from many tests.
141
def __init__(self, stream, descriptions, verbosity,
145
"""Construct new TestResult.
147
:param bench_history: Optionally, a writable file object to accumulate
150
unittest._TextTestResult.__init__(self, stream, descriptions, verbosity)
151
if bench_history is not None:
152
from bzrlib.version import _get_bzr_source_tree
153
src_tree = _get_bzr_source_tree()
156
revision_id = src_tree.get_parent_ids()[0]
158
# XXX: if this is a brand new tree, do the same as if there
162
# XXX: If there's no branch, what should we do?
164
bench_history.write("--date %s %s\n" % (time.time(), revision_id))
165
self._bench_history = bench_history
166
self.ui = ui.ui_factory
169
self.failure_count = 0
170
self.known_failure_count = 0
172
self.not_applicable_count = 0
173
self.unsupported = {}
175
self._overall_start_time = time.time()
176
self._strict = strict
179
# nb: called stopTestRun in the version of this that Python merged
180
# upstream, according to lifeless 20090803
182
ok = self.wasStrictlySuccessful()
184
ok = self.wasSuccessful()
186
self.stream.write('tests passed\n')
188
self.stream.write('tests failed\n')
189
if TestCase._first_thread_leaker_id:
191
'%s is leaking threads among %d leaking tests.\n' % (
192
TestCase._first_thread_leaker_id,
193
TestCase._leaking_threads_tests))
195
def _extractBenchmarkTime(self, testCase):
196
"""Add a benchmark time for the current test case."""
197
return getattr(testCase, "_benchtime", None)
199
def _elapsedTestTimeString(self):
200
"""Return a time string for the overall time the current test has taken."""
201
return self._formatTime(time.time() - self._start_time)
203
def _testTimeString(self, testCase):
204
benchmark_time = self._extractBenchmarkTime(testCase)
205
if benchmark_time is not None:
206
return self._formatTime(benchmark_time) + "*"
208
return self._elapsedTestTimeString()
210
def _formatTime(self, seconds):
211
"""Format seconds as milliseconds with leading spaces."""
212
# some benchmarks can take thousands of seconds to run, so we need 8
214
return "%8dms" % (1000 * seconds)
216
def _shortened_test_description(self, test):
218
what = re.sub(r'^bzrlib\.(tests|benchmarks)\.', '', what)
221
def startTest(self, test):
222
unittest.TestResult.startTest(self, test)
225
self.report_test_start(test)
226
test.number = self.count
227
self._recordTestStartTime()
229
def startTests(self):
231
if getattr(sys, 'frozen', None) is None:
232
bzr_path = osutils.realpath(sys.argv[0])
234
bzr_path = sys.executable
236
'testing: %s\n' % (bzr_path,))
239
bzrlib.__path__[0],))
241
' bzr-%s python-%s %s\n' % (
242
bzrlib.version_string,
243
bzrlib._format_version_tuple(sys.version_info),
244
platform.platform(aliased=1),
246
self.stream.write('\n')
248
def _recordTestStartTime(self):
249
"""Record that a test has started."""
250
self._start_time = time.time()
252
def _cleanupLogFile(self, test):
253
# We can only do this if we have one of our TestCases, not if
255
setKeepLogfile = getattr(test, 'setKeepLogfile', None)
256
if setKeepLogfile is not None:
259
def addError(self, test, err):
260
"""Tell result that test finished with an error.
262
Called from the TestCase run() method when the test
263
fails with an unexpected error.
265
self._testConcluded(test)
266
if isinstance(err[1], TestNotApplicable):
267
return self._addNotApplicable(test, err)
268
elif isinstance(err[1], UnavailableFeature):
269
return self.addNotSupported(test, err[1].args[0])
271
unittest.TestResult.addError(self, test, err)
272
self.error_count += 1
273
self.report_error(test, err)
276
self._cleanupLogFile(test)
278
def addFailure(self, test, err):
279
"""Tell result that test failed.
281
Called from the TestCase run() method when the test
282
fails because e.g. an assert() method failed.
284
self._testConcluded(test)
285
if isinstance(err[1], KnownFailure):
286
return self._addKnownFailure(test, err)
288
unittest.TestResult.addFailure(self, test, err)
289
self.failure_count += 1
290
self.report_failure(test, err)
293
self._cleanupLogFile(test)
295
def addSuccess(self, test):
296
"""Tell result that test completed successfully.
298
Called from the TestCase run()
300
self._testConcluded(test)
301
if self._bench_history is not None:
302
benchmark_time = self._extractBenchmarkTime(test)
303
if benchmark_time is not None:
304
self._bench_history.write("%s %s\n" % (
305
self._formatTime(benchmark_time),
307
self.report_success(test)
308
self._cleanupLogFile(test)
309
unittest.TestResult.addSuccess(self, test)
310
test._log_contents = ''
312
def _testConcluded(self, test):
313
"""Common code when a test has finished.
315
Called regardless of whether it succeded, failed, etc.
319
def _addKnownFailure(self, test, err):
320
self.known_failure_count += 1
321
self.report_known_failure(test, err)
323
def addNotSupported(self, test, feature):
324
"""The test will not be run because of a missing feature.
326
# this can be called in two different ways: it may be that the
327
# test started running, and then raised (through addError)
328
# UnavailableFeature. Alternatively this method can be called
329
# while probing for features before running the tests; in that
330
# case we will see startTest and stopTest, but the test will never
332
self.unsupported.setdefault(str(feature), 0)
333
self.unsupported[str(feature)] += 1
334
self.report_unsupported(test, feature)
336
def addSkip(self, test, reason):
337
"""A test has not run for 'reason'."""
339
self.report_skip(test, reason)
341
def _addNotApplicable(self, test, skip_excinfo):
342
if isinstance(skip_excinfo[1], TestNotApplicable):
343
self.not_applicable_count += 1
344
self.report_not_applicable(test, skip_excinfo)
347
except KeyboardInterrupt:
350
self.addError(test, test.exc_info())
352
# seems best to treat this as success from point-of-view of unittest
353
# -- it actually does nothing so it barely matters :)
354
unittest.TestResult.addSuccess(self, test)
355
test._log_contents = ''
357
def printErrorList(self, flavour, errors):
358
for test, err in errors:
359
self.stream.writeln(self.separator1)
360
self.stream.write("%s: " % flavour)
361
self.stream.writeln(self.getDescription(test))
362
if getattr(test, '_get_log', None) is not None:
363
log_contents = test._get_log()
365
self.stream.write('\n')
367
('vvvv[log from %s]' % test.id()).ljust(78,'-'))
368
self.stream.write('\n')
369
self.stream.write(log_contents)
370
self.stream.write('\n')
372
('^^^^[log from %s]' % test.id()).ljust(78,'-'))
373
self.stream.write('\n')
374
self.stream.writeln(self.separator2)
375
self.stream.writeln("%s" % err)
377
def progress(self, offset, whence):
378
"""The test is adjusting the count of tests to run."""
379
if whence == SUBUNIT_SEEK_SET:
380
self.num_tests = offset
381
elif whence == SUBUNIT_SEEK_CUR:
382
self.num_tests += offset
384
raise errors.BzrError("Unknown whence %r" % whence)
389
def report_cleaning_up(self):
392
def report_success(self, test):
395
def wasStrictlySuccessful(self):
396
if self.unsupported or self.known_failure_count:
398
return self.wasSuccessful()
401
class TextTestResult(ExtendedTestResult):
402
"""Displays progress and results of tests in text form"""
404
def __init__(self, stream, descriptions, verbosity,
409
ExtendedTestResult.__init__(self, stream, descriptions, verbosity,
410
bench_history, strict)
411
# We no longer pass them around, but just rely on the UIFactory stack
414
warnings.warn("Passing pb to TextTestResult is deprecated")
415
self.pb = self.ui.nested_progress_bar()
416
self.pb.show_pct = False
417
self.pb.show_spinner = False
418
self.pb.show_eta = False,
419
self.pb.show_count = False
420
self.pb.show_bar = False
421
self.pb.update_latency = 0
422
self.pb.show_transport_activity = False
425
# called when the tests that are going to run have run
427
super(TextTestResult, self).done()
432
def report_starting(self):
433
self.pb.update('[test 0/%d] Starting' % (self.num_tests))
435
def printErrors(self):
436
# clear the pb to make room for the error listing
438
super(TextTestResult, self).printErrors()
440
def _progress_prefix_text(self):
441
# the longer this text, the less space we have to show the test
443
a = '[%d' % self.count # total that have been run
444
# tests skipped as known not to be relevant are not important enough
446
## if self.skip_count:
447
## a += ', %d skip' % self.skip_count
448
## if self.known_failure_count:
449
## a += '+%dX' % self.known_failure_count
451
a +='/%d' % self.num_tests
453
runtime = time.time() - self._overall_start_time
455
a += '%dm%ds' % (runtime / 60, runtime % 60)
459
a += ', %d err' % self.error_count
460
if self.failure_count:
461
a += ', %d fail' % self.failure_count
463
a += ', %d missing' % len(self.unsupported)
467
def report_test_start(self, test):
470
self._progress_prefix_text()
472
+ self._shortened_test_description(test))
474
def _test_description(self, test):
475
return self._shortened_test_description(test)
477
def report_error(self, test, err):
478
self.pb.note('ERROR: %s\n %s\n',
479
self._test_description(test),
483
def report_failure(self, test, err):
484
self.pb.note('FAIL: %s\n %s\n',
485
self._test_description(test),
489
def report_known_failure(self, test, err):
490
self.pb.note('XFAIL: %s\n%s\n',
491
self._test_description(test), err[1])
493
def report_skip(self, test, reason):
496
def report_not_applicable(self, test, skip_excinfo):
499
def report_unsupported(self, test, feature):
500
"""test cannot be run because feature is missing."""
502
def report_cleaning_up(self):
503
self.pb.update('Cleaning up')
506
class VerboseTestResult(ExtendedTestResult):
507
"""Produce long output, with one line per test run plus times"""
509
def _ellipsize_to_right(self, a_string, final_width):
510
"""Truncate and pad a string, keeping the right hand side"""
511
if len(a_string) > final_width:
512
result = '...' + a_string[3-final_width:]
515
return result.ljust(final_width)
517
def report_starting(self):
518
self.stream.write('running %d tests...\n' % self.num_tests)
520
def report_test_start(self, test):
522
name = self._shortened_test_description(test)
523
# width needs space for 6 char status, plus 1 for slash, plus an
524
# 11-char time string, plus a trailing blank
525
# when NUMBERED_DIRS: plus 5 chars on test number, plus 1 char on space
526
self.stream.write(self._ellipsize_to_right(name,
527
osutils.terminal_width()-18))
530
def _error_summary(self, err):
532
return '%s%s' % (indent, err[1])
534
def report_error(self, test, err):
535
self.stream.writeln('ERROR %s\n%s'
536
% (self._testTimeString(test),
537
self._error_summary(err)))
539
def report_failure(self, test, err):
540
self.stream.writeln(' FAIL %s\n%s'
541
% (self._testTimeString(test),
542
self._error_summary(err)))
544
def report_known_failure(self, test, err):
545
self.stream.writeln('XFAIL %s\n%s'
546
% (self._testTimeString(test),
547
self._error_summary(err)))
549
def report_success(self, test):
550
self.stream.writeln(' OK %s' % self._testTimeString(test))
551
for bench_called, stats in getattr(test, '_benchcalls', []):
552
self.stream.writeln('LSProf output for %s(%s, %s)' % bench_called)
553
stats.pprint(file=self.stream)
554
# flush the stream so that we get smooth output. This verbose mode is
555
# used to show the output in PQM.
558
def report_skip(self, test, reason):
559
self.stream.writeln(' SKIP %s\n%s'
560
% (self._testTimeString(test), reason))
562
def report_not_applicable(self, test, skip_excinfo):
563
self.stream.writeln(' N/A %s\n%s'
564
% (self._testTimeString(test),
565
self._error_summary(skip_excinfo)))
567
def report_unsupported(self, test, feature):
568
"""test cannot be run because feature is missing."""
569
self.stream.writeln("NODEP %s\n The feature '%s' is not available."
570
%(self._testTimeString(test), feature))
573
class TextTestRunner(object):
574
stop_on_failure = False
584
self.stream = unittest._WritelnDecorator(stream)
585
self.descriptions = descriptions
586
self.verbosity = verbosity
587
self._bench_history = bench_history
588
self.list_only = list_only
589
self._strict = strict
592
"Run the given test case or test suite."
593
startTime = time.time()
594
if self.verbosity == 1:
595
result_class = TextTestResult
596
elif self.verbosity >= 2:
597
result_class = VerboseTestResult
598
result = result_class(self.stream,
601
bench_history=self._bench_history,
604
result.stop_early = self.stop_on_failure
605
result.report_starting()
607
if self.verbosity >= 2:
608
self.stream.writeln("Listing tests only ...\n")
610
for t in iter_suite_tests(test):
611
self.stream.writeln("%s" % (t.id()))
620
if isinstance(test, testtools.ConcurrentTestSuite):
621
# We need to catch bzr specific behaviors
622
test.run(BZRTransformingResult(result))
625
run = result.testsRun
627
stopTime = time.time()
628
timeTaken = stopTime - startTime
630
self.stream.writeln(result.separator2)
631
self.stream.writeln("%s %d test%s in %.3fs" % (actionTaken,
632
run, run != 1 and "s" or "", timeTaken))
633
self.stream.writeln()
634
if not result.wasSuccessful():
635
self.stream.write("FAILED (")
636
failed, errored = map(len, (result.failures, result.errors))
638
self.stream.write("failures=%d" % failed)
640
if failed: self.stream.write(", ")
641
self.stream.write("errors=%d" % errored)
642
if result.known_failure_count:
643
if failed or errored: self.stream.write(", ")
644
self.stream.write("known_failure_count=%d" %
645
result.known_failure_count)
646
self.stream.writeln(")")
648
if result.known_failure_count:
649
self.stream.writeln("OK (known_failures=%d)" %
650
result.known_failure_count)
652
self.stream.writeln("OK")
653
if result.skip_count > 0:
654
skipped = result.skip_count
655
self.stream.writeln('%d test%s skipped' %
656
(skipped, skipped != 1 and "s" or ""))
657
if result.unsupported:
658
for feature, count in sorted(result.unsupported.items()):
659
self.stream.writeln("Missing feature '%s' skipped %d tests." %
665
def iter_suite_tests(suite):
666
"""Return all tests in a suite, recursing through nested suites"""
667
if isinstance(suite, unittest.TestCase):
669
elif isinstance(suite, unittest.TestSuite):
671
for r in iter_suite_tests(item):
674
raise Exception('unknown type %r for object %r'
675
% (type(suite), suite))
678
class TestSkipped(Exception):
679
"""Indicates that a test was intentionally skipped, rather than failing."""
682
class TestNotApplicable(TestSkipped):
683
"""A test is not applicable to the situation where it was run.
685
This is only normally raised by parameterized tests, if they find that
686
the instance they're constructed upon does not support one aspect
691
class KnownFailure(AssertionError):
692
"""Indicates that a test failed in a precisely expected manner.
694
Such failures dont block the whole test suite from passing because they are
695
indicators of partially completed code or of future work. We have an
696
explicit error for them so that we can ensure that they are always visible:
697
KnownFailures are always shown in the output of bzr selftest.
701
class UnavailableFeature(Exception):
702
"""A feature required for this test was not available.
704
The feature should be used to construct the exception.
29
708
class CommandFailed(Exception):
33
class TestBase(TestCase):
34
"""Base class for bzr test cases.
36
Just defines some useful helper functions; doesn't actually test
40
# TODO: Special methods to invoke bzr, so that we can run it
41
# through a specified Python intepreter
43
OVERRIDE_PYTHON = None # to run with alternative python 'python'
49
def formcmd(self, cmd):
50
if isinstance(cmd, basestring):
55
if self.OVERRIDE_PYTHON:
56
cmd.insert(0, self.OVERRIDE_PYTHON)
58
self.log('$ %r' % cmd)
63
def runcmd(self, cmd, retcode=0):
64
"""Run one command and check the return code.
66
Returns a tuple of (stdout,stderr) strings.
68
If a single string is based, it is split into words.
69
For commands that are not simple space-separated words, please
70
pass a list instead."""
71
cmd = self.formcmd(cmd)
73
self.log('$ ' + ' '.join(cmd))
74
actual_retcode = call(cmd, stdout=self.TEST_LOG, stderr=self.TEST_LOG)
76
if retcode != actual_retcode:
77
raise CommandFailed("test failed: %r returned %d, expected %d"
78
% (cmd, actual_retcode, retcode))
81
def backtick(self, cmd, retcode=0):
82
"""Run a command and return its output"""
83
cmd = self.formcmd(cmd)
84
child = Popen(cmd, stdout=PIPE, stderr=self.TEST_LOG)
85
outd, errd = child.communicate()
87
actual_retcode = child.wait()
89
outd = outd.replace('\r', '')
91
if retcode != actual_retcode:
92
raise CommandFailed("test failed: %r returned %d, expected %d"
93
% (cmd, actual_retcode, retcode))
99
def build_tree(self, shape):
100
"""Build a test tree according to a pattern.
102
shape is a sequence of file specifications. If the final
103
character is '/', a directory is created.
105
This doesn't add anything to a branch.
107
# XXX: It's OK to just create them using forward slashes on windows?
110
assert isinstance(name, basestring)
115
print >>f, "contents of", name
120
"""Log a message to a progress file"""
121
self._log_buf = self._log_buf + str(msg) + '\n'
122
print >>self.TEST_LOG, msg
712
class StringIOWrapper(object):
713
"""A wrapper around cStringIO which just adds an encoding attribute.
715
Internally we can check sys.stdout to see what the output encoding
716
should be. However, cStringIO has no encoding attribute that we can
717
set. So we wrap it instead.
722
def __init__(self, s=None):
724
self.__dict__['_cstring'] = StringIO(s)
726
self.__dict__['_cstring'] = StringIO()
728
def __getattr__(self, name, getattr=getattr):
729
return getattr(self.__dict__['_cstring'], name)
731
def __setattr__(self, name, val):
732
if name == 'encoding':
733
self.__dict__['encoding'] = val
735
return setattr(self._cstring, name, val)
738
class TestUIFactory(TextUIFactory):
739
"""A UI Factory for testing.
741
Hide the progress bar but emit note()s.
743
Allows get_password to be tested without real tty attached.
745
See also CannedInputUIFactory which lets you provide programmatic input in
748
# TODO: Capture progress events at the model level and allow them to be
749
# observed by tests that care.
751
# XXX: Should probably unify more with CannedInputUIFactory or a
752
# particular configuration of TextUIFactory, or otherwise have a clearer
753
# idea of how they're supposed to be different.
754
# See https://bugs.edge.launchpad.net/bzr/+bug/408213
756
def __init__(self, stdout=None, stderr=None, stdin=None):
757
if stdin is not None:
758
# We use a StringIOWrapper to be able to test various
759
# encodings, but the user is still responsible to
760
# encode the string and to set the encoding attribute
761
# of StringIOWrapper.
762
stdin = StringIOWrapper(stdin)
763
super(TestUIFactory, self).__init__(stdin, stdout, stderr)
765
def get_non_echoed_password(self):
766
"""Get password from stdin without trying to handle the echo mode"""
767
password = self.stdin.readline()
770
if password[-1] == '\n':
771
password = password[:-1]
774
def make_progress_view(self):
775
return NullProgressView()
778
class TestCase(unittest.TestCase):
779
"""Base class for bzr unit tests.
781
Tests that need access to disk resources should subclass
782
TestCaseInTempDir not TestCase.
784
Error and debug log messages are redirected from their usual
785
location into a temporary file, the contents of which can be
786
retrieved by _get_log(). We use a real OS file, not an in-memory object,
787
so that it can also capture file IO. When the test completes this file
788
is read into memory and removed from disk.
790
There are also convenience functions to invoke bzr's command-line
791
routine, and to build and check bzr trees.
793
In addition to the usual method of overriding tearDown(), this class also
794
allows subclasses to register functions into the _cleanups list, which is
795
run in order as the object is torn down. It's less likely this will be
796
accidentally overlooked.
799
_active_threads = None
800
_leaking_threads_tests = 0
801
_first_thread_leaker_id = None
802
_log_file_name = None
804
_keep_log_file = False
805
# record lsprof data when performing benchmark calls.
806
_gather_lsprof_in_benchmarks = False
807
attrs_to_keep = ('id', '_testMethodName', '_testMethodDoc',
808
'_log_contents', '_log_file_name', '_benchtime',
809
'_TestCase__testMethodName', '_TestCase__testMethodDoc',)
811
def __init__(self, methodName='testMethod'):
812
super(TestCase, self).__init__(methodName)
814
self._bzr_test_setUp_run = False
815
self._bzr_test_tearDown_run = False
818
unittest.TestCase.setUp(self)
819
self._bzr_test_setUp_run = True
820
self._cleanEnvironment()
823
self._benchcalls = []
824
self._benchtime = None
827
self._clear_debug_flags()
828
TestCase._active_threads = threading.activeCount()
829
self.addCleanup(self._check_leaked_threads)
834
pdb.Pdb().set_trace(sys._getframe().f_back)
836
def _check_leaked_threads(self):
837
active = threading.activeCount()
838
leaked_threads = active - TestCase._active_threads
839
TestCase._active_threads = active
841
TestCase._leaking_threads_tests += 1
842
if TestCase._first_thread_leaker_id is None:
843
TestCase._first_thread_leaker_id = self.id()
845
def _clear_debug_flags(self):
846
"""Prevent externally set debug flags affecting tests.
848
Tests that want to use debug flags can just set them in the
849
debug_flags set during setup/teardown.
851
self._preserved_debug_flags = set(debug.debug_flags)
852
if 'allow_debug' not in selftest_debug_flags:
853
debug.debug_flags.clear()
854
if 'disable_lock_checks' not in selftest_debug_flags:
855
debug.debug_flags.add('strict_locks')
856
self.addCleanup(self._restore_debug_flags)
858
def _clear_hooks(self):
859
# prevent hooks affecting tests
860
self._preserved_hooks = {}
861
for key, factory in hooks.known_hooks.items():
862
parent, name = hooks.known_hooks_key_to_parent_and_attribute(key)
863
current_hooks = hooks.known_hooks_key_to_object(key)
864
self._preserved_hooks[parent] = (name, current_hooks)
865
self.addCleanup(self._restoreHooks)
866
for key, factory in hooks.known_hooks.items():
867
parent, name = hooks.known_hooks_key_to_parent_and_attribute(key)
868
setattr(parent, name, factory())
869
# this hook should always be installed
870
request._install_hook()
872
def _silenceUI(self):
873
"""Turn off UI for duration of test"""
874
# by default the UI is off; tests can turn it on if they want it.
875
saved = ui.ui_factory
877
ui.ui_factory = saved
878
ui.ui_factory = ui.SilentUIFactory()
879
self.addCleanup(_restore)
881
def _check_locks(self):
882
"""Check that all lock take/release actions have been paired."""
883
# We always check for mismatched locks. If a mismatch is found, we
884
# fail unless -Edisable_lock_checks is supplied to selftest, in which
885
# case we just print a warning.
887
acquired_locks = [lock for action, lock in self._lock_actions
888
if action == 'acquired']
889
released_locks = [lock for action, lock in self._lock_actions
890
if action == 'released']
891
broken_locks = [lock for action, lock in self._lock_actions
892
if action == 'broken']
893
# trivially, given the tests for lock acquistion and release, if we
894
# have as many in each list, it should be ok. Some lock tests also
895
# break some locks on purpose and should be taken into account by
896
# considering that breaking a lock is just a dirty way of releasing it.
897
if len(acquired_locks) != (len(released_locks) + len(broken_locks)):
898
message = ('Different number of acquired and '
899
'released or broken locks. (%s, %s + %s)' %
900
(acquired_locks, released_locks, broken_locks))
901
if not self._lock_check_thorough:
902
# Rather than fail, just warn
903
print "Broken test %s: %s" % (self, message)
907
def _track_locks(self):
908
"""Track lock activity during tests."""
909
self._lock_actions = []
910
if 'disable_lock_checks' in selftest_debug_flags:
911
self._lock_check_thorough = False
913
self._lock_check_thorough = True
915
self.addCleanup(self._check_locks)
916
_mod_lock.Lock.hooks.install_named_hook('lock_acquired',
917
self._lock_acquired, None)
918
_mod_lock.Lock.hooks.install_named_hook('lock_released',
919
self._lock_released, None)
920
_mod_lock.Lock.hooks.install_named_hook('lock_broken',
921
self._lock_broken, None)
923
def _lock_acquired(self, result):
924
self._lock_actions.append(('acquired', result))
926
def _lock_released(self, result):
927
self._lock_actions.append(('released', result))
929
def _lock_broken(self, result):
930
self._lock_actions.append(('broken', result))
932
def _ndiff_strings(self, a, b):
933
"""Return ndiff between two strings containing lines.
935
A trailing newline is added if missing to make the strings
937
if b and b[-1] != '\n':
939
if a and a[-1] != '\n':
941
difflines = difflib.ndiff(a.splitlines(True),
943
linejunk=lambda x: False,
944
charjunk=lambda x: False)
945
return ''.join(difflines)
947
def assertEqual(self, a, b, message=''):
951
except UnicodeError, e:
952
# If we can't compare without getting a UnicodeError, then
953
# obviously they are different
954
mutter('UnicodeError: %s', e)
957
raise AssertionError("%snot equal:\na = %s\nb = %s\n"
959
pformat(a), pformat(b)))
961
assertEquals = assertEqual
963
def assertEqualDiff(self, a, b, message=None):
964
"""Assert two texts are equal, if not raise an exception.
966
This is intended for use with multi-line strings where it can
967
be hard to find the differences by eye.
969
# TODO: perhaps override assertEquals to call this for strings?
973
message = "texts not equal:\n"
975
message = 'first string is missing a final newline.\n'
977
message = 'second string is missing a final newline.\n'
978
raise AssertionError(message +
979
self._ndiff_strings(a, b))
981
def assertEqualMode(self, mode, mode_test):
982
self.assertEqual(mode, mode_test,
983
'mode mismatch %o != %o' % (mode, mode_test))
985
def assertEqualStat(self, expected, actual):
986
"""assert that expected and actual are the same stat result.
988
:param expected: A stat result.
989
:param actual: A stat result.
990
:raises AssertionError: If the expected and actual stat values differ
993
self.assertEqual(expected.st_size, actual.st_size)
994
self.assertEqual(expected.st_mtime, actual.st_mtime)
995
self.assertEqual(expected.st_ctime, actual.st_ctime)
996
self.assertEqual(expected.st_dev, actual.st_dev)
997
self.assertEqual(expected.st_ino, actual.st_ino)
998
self.assertEqual(expected.st_mode, actual.st_mode)
1000
def assertLength(self, length, obj_with_len):
1001
"""Assert that obj_with_len is of length length."""
1002
if len(obj_with_len) != length:
1003
self.fail("Incorrect length: wanted %d, got %d for %r" % (
1004
length, len(obj_with_len), obj_with_len))
1006
def assertPositive(self, val):
1007
"""Assert that val is greater than 0."""
1008
self.assertTrue(val > 0, 'expected a positive value, but got %s' % val)
1010
def assertNegative(self, val):
1011
"""Assert that val is less than 0."""
1012
self.assertTrue(val < 0, 'expected a negative value, but got %s' % val)
1014
def assertStartsWith(self, s, prefix):
1015
if not s.startswith(prefix):
1016
raise AssertionError('string %r does not start with %r' % (s, prefix))
1018
def assertEndsWith(self, s, suffix):
1019
"""Asserts that s ends with suffix."""
1020
if not s.endswith(suffix):
1021
raise AssertionError('string %r does not end with %r' % (s, suffix))
1023
def assertContainsRe(self, haystack, needle_re, flags=0):
1024
"""Assert that a contains something matching a regular expression."""
1025
if not re.search(needle_re, haystack, flags):
1026
if '\n' in haystack or len(haystack) > 60:
1027
# a long string, format it in a more readable way
1028
raise AssertionError(
1029
'pattern "%s" not found in\n"""\\\n%s"""\n'
1030
% (needle_re, haystack))
1032
raise AssertionError('pattern "%s" not found in "%s"'
1033
% (needle_re, haystack))
1035
def assertNotContainsRe(self, haystack, needle_re, flags=0):
1036
"""Assert that a does not match a regular expression"""
1037
if re.search(needle_re, haystack, flags):
1038
raise AssertionError('pattern "%s" found in "%s"'
1039
% (needle_re, haystack))
1041
def assertSubset(self, sublist, superlist):
1042
"""Assert that every entry in sublist is present in superlist."""
1043
missing = set(sublist) - set(superlist)
1044
if len(missing) > 0:
1045
raise AssertionError("value(s) %r not present in container %r" %
1046
(missing, superlist))
1048
def assertListRaises(self, excClass, func, *args, **kwargs):
1049
"""Fail unless excClass is raised when the iterator from func is used.
1051
Many functions can return generators this makes sure
1052
to wrap them in a list() call to make sure the whole generator
1053
is run, and that the proper exception is raised.
1056
list(func(*args, **kwargs))
1060
if getattr(excClass,'__name__', None) is not None:
1061
excName = excClass.__name__
1063
excName = str(excClass)
1064
raise self.failureException, "%s not raised" % excName
1066
def assertRaises(self, excClass, callableObj, *args, **kwargs):
1067
"""Assert that a callable raises a particular exception.
1069
:param excClass: As for the except statement, this may be either an
1070
exception class, or a tuple of classes.
1071
:param callableObj: A callable, will be passed ``*args`` and
1074
Returns the exception so that you can examine it.
1077
callableObj(*args, **kwargs)
1081
if getattr(excClass,'__name__', None) is not None:
1082
excName = excClass.__name__
1085
excName = str(excClass)
1086
raise self.failureException, "%s not raised" % excName
1088
def assertIs(self, left, right, message=None):
1089
if not (left is right):
1090
if message is not None:
1091
raise AssertionError(message)
1093
raise AssertionError("%r is not %r." % (left, right))
1095
def assertIsNot(self, left, right, message=None):
1097
if message is not None:
1098
raise AssertionError(message)
1100
raise AssertionError("%r is %r." % (left, right))
1102
def assertTransportMode(self, transport, path, mode):
1103
"""Fail if a path does not have mode "mode".
1105
If modes are not supported on this transport, the assertion is ignored.
1107
if not transport._can_roundtrip_unix_modebits():
1109
path_stat = transport.stat(path)
1110
actual_mode = stat.S_IMODE(path_stat.st_mode)
1111
self.assertEqual(mode, actual_mode,
1112
'mode of %r incorrect (%s != %s)'
1113
% (path, oct(mode), oct(actual_mode)))
1115
def assertIsSameRealPath(self, path1, path2):
1116
"""Fail if path1 and path2 points to different files"""
1117
self.assertEqual(osutils.realpath(path1),
1118
osutils.realpath(path2),
1119
"apparent paths:\na = %s\nb = %s\n," % (path1, path2))
1121
def assertIsInstance(self, obj, kls, msg=None):
1122
"""Fail if obj is not an instance of kls
1124
:param msg: Supplementary message to show if the assertion fails.
1126
if not isinstance(obj, kls):
1127
m = "%r is an instance of %s rather than %s" % (
1128
obj, obj.__class__, kls)
1133
def expectFailure(self, reason, assertion, *args, **kwargs):
1134
"""Invoke a test, expecting it to fail for the given reason.
1136
This is for assertions that ought to succeed, but currently fail.
1137
(The failure is *expected* but not *wanted*.) Please be very precise
1138
about the failure you're expecting. If a new bug is introduced,
1139
AssertionError should be raised, not KnownFailure.
1141
Frequently, expectFailure should be followed by an opposite assertion.
1144
Intended to be used with a callable that raises AssertionError as the
1145
'assertion' parameter. args and kwargs are passed to the 'assertion'.
1147
Raises KnownFailure if the test fails. Raises AssertionError if the
1152
self.expectFailure('Math is broken', self.assertNotEqual, 54,
1154
self.assertEqual(42, dynamic_val)
1156
This means that a dynamic_val of 54 will cause the test to raise
1157
a KnownFailure. Once math is fixed and the expectFailure is removed,
1158
only a dynamic_val of 42 will allow the test to pass. Anything other
1159
than 54 or 42 will cause an AssertionError.
1162
assertion(*args, **kwargs)
1163
except AssertionError:
1164
raise KnownFailure(reason)
1166
self.fail('Unexpected success. Should have failed: %s' % reason)
1168
def assertFileEqual(self, content, path):
1169
"""Fail if path does not contain 'content'."""
1170
self.failUnlessExists(path)
1171
f = file(path, 'rb')
1176
self.assertEqualDiff(content, s)
1178
def failUnlessExists(self, path):
1179
"""Fail unless path or paths, which may be abs or relative, exist."""
1180
if not isinstance(path, basestring):
1182
self.failUnlessExists(p)
1184
self.failUnless(osutils.lexists(path),path+" does not exist")
1186
def failIfExists(self, path):
1187
"""Fail if path or paths, which may be abs or relative, exist."""
1188
if not isinstance(path, basestring):
1190
self.failIfExists(p)
1192
self.failIf(osutils.lexists(path),path+" exists")
1194
def _capture_deprecation_warnings(self, a_callable, *args, **kwargs):
1195
"""A helper for callDeprecated and applyDeprecated.
1197
:param a_callable: A callable to call.
1198
:param args: The positional arguments for the callable
1199
:param kwargs: The keyword arguments for the callable
1200
:return: A tuple (warnings, result). result is the result of calling
1201
a_callable(``*args``, ``**kwargs``).
1204
def capture_warnings(msg, cls=None, stacklevel=None):
1205
# we've hooked into a deprecation specific callpath,
1206
# only deprecations should getting sent via it.
1207
self.assertEqual(cls, DeprecationWarning)
1208
local_warnings.append(msg)
1209
original_warning_method = symbol_versioning.warn
1210
symbol_versioning.set_warning_method(capture_warnings)
1212
result = a_callable(*args, **kwargs)
1214
symbol_versioning.set_warning_method(original_warning_method)
1215
return (local_warnings, result)
1217
def applyDeprecated(self, deprecation_format, a_callable, *args, **kwargs):
1218
"""Call a deprecated callable without warning the user.
1220
Note that this only captures warnings raised by symbol_versioning.warn,
1221
not other callers that go direct to the warning module.
1223
To test that a deprecated method raises an error, do something like
1226
self.assertRaises(errors.ReservedId,
1227
self.applyDeprecated,
1228
deprecated_in((1, 5, 0)),
1232
:param deprecation_format: The deprecation format that the callable
1233
should have been deprecated with. This is the same type as the
1234
parameter to deprecated_method/deprecated_function. If the
1235
callable is not deprecated with this format, an assertion error
1237
:param a_callable: A callable to call. This may be a bound method or
1238
a regular function. It will be called with ``*args`` and
1240
:param args: The positional arguments for the callable
1241
:param kwargs: The keyword arguments for the callable
1242
:return: The result of a_callable(``*args``, ``**kwargs``)
1244
call_warnings, result = self._capture_deprecation_warnings(a_callable,
1246
expected_first_warning = symbol_versioning.deprecation_string(
1247
a_callable, deprecation_format)
1248
if len(call_warnings) == 0:
1249
self.fail("No deprecation warning generated by call to %s" %
1251
self.assertEqual(expected_first_warning, call_warnings[0])
1254
def callCatchWarnings(self, fn, *args, **kw):
1255
"""Call a callable that raises python warnings.
1257
The caller's responsible for examining the returned warnings.
1259
If the callable raises an exception, the exception is not
1260
caught and propagates up to the caller. In that case, the list
1261
of warnings is not available.
1263
:returns: ([warning_object, ...], fn_result)
1265
# XXX: This is not perfect, because it completely overrides the
1266
# warnings filters, and some code may depend on suppressing particular
1267
# warnings. It's the easiest way to insulate ourselves from -Werror,
1268
# though. -- Andrew, 20071062
1270
def _catcher(message, category, filename, lineno, file=None, line=None):
1271
# despite the name, 'message' is normally(?) a Warning subclass
1273
wlist.append(message)
1274
saved_showwarning = warnings.showwarning
1275
saved_filters = warnings.filters
1277
warnings.showwarning = _catcher
1278
warnings.filters = []
1279
result = fn(*args, **kw)
1281
warnings.showwarning = saved_showwarning
1282
warnings.filters = saved_filters
1283
return wlist, result
1285
def callDeprecated(self, expected, callable, *args, **kwargs):
1286
"""Assert that a callable is deprecated in a particular way.
1288
This is a very precise test for unusual requirements. The
1289
applyDeprecated helper function is probably more suited for most tests
1290
as it allows you to simply specify the deprecation format being used
1291
and will ensure that that is issued for the function being called.
1293
Note that this only captures warnings raised by symbol_versioning.warn,
1294
not other callers that go direct to the warning module. To catch
1295
general warnings, use callCatchWarnings.
1297
:param expected: a list of the deprecation warnings expected, in order
1298
:param callable: The callable to call
1299
:param args: The positional arguments for the callable
1300
:param kwargs: The keyword arguments for the callable
1302
call_warnings, result = self._capture_deprecation_warnings(callable,
1304
self.assertEqual(expected, call_warnings)
1307
def _startLogFile(self):
1308
"""Send bzr and test log messages to a temporary file.
1310
The file is removed as the test is torn down.
1312
fileno, name = tempfile.mkstemp(suffix='.log', prefix='testbzr')
1313
self._log_file = os.fdopen(fileno, 'w+')
1314
self._log_memento = bzrlib.trace.push_log_file(self._log_file)
1315
self._log_file_name = name
1316
self.addCleanup(self._finishLogFile)
1318
def _finishLogFile(self):
1319
"""Finished with the log file.
1321
Close the file and delete it, unless setKeepLogfile was called.
1323
if self._log_file is None:
1325
bzrlib.trace.pop_log_file(self._log_memento)
1326
self._log_file.close()
1327
self._log_file = None
1328
if not self._keep_log_file:
1329
os.remove(self._log_file_name)
1330
self._log_file_name = None
1332
def setKeepLogfile(self):
1333
"""Make the logfile not be deleted when _finishLogFile is called."""
1334
self._keep_log_file = True
1336
def thisFailsStrictLockCheck(self):
1337
"""It is known that this test would fail with -Dstrict_locks.
1339
By default, all tests are run with strict lock checking unless
1340
-Edisable_lock_checks is supplied. However there are some tests which
1341
we know fail strict locks at this point that have not been fixed.
1342
They should call this function to disable the strict checking.
1344
This should be used sparingly, it is much better to fix the locking
1345
issues rather than papering over the problem by calling this function.
1347
debug.debug_flags.discard('strict_locks')
1349
def addCleanup(self, callable, *args, **kwargs):
1350
"""Arrange to run a callable when this case is torn down.
1352
Callables are run in the reverse of the order they are registered,
1353
ie last-in first-out.
1355
self._cleanups.append((callable, args, kwargs))
1357
def _cleanEnvironment(self):
1359
'BZR_HOME': None, # Don't inherit BZR_HOME to all the tests.
1360
'HOME': os.getcwd(),
1361
# bzr now uses the Win32 API and doesn't rely on APPDATA, but the
1362
# tests do check our impls match APPDATA
1363
'BZR_EDITOR': None, # test_msgeditor manipulates this variable
1367
'BZREMAIL': None, # may still be present in the environment
1369
'BZR_PROGRESS_BAR': None,
1371
'BZR_PLUGIN_PATH': None,
1372
# Make sure that any text ui tests are consistent regardless of
1373
# the environment the test case is run in; you may want tests that
1374
# test other combinations. 'dumb' is a reasonable guess for tests
1375
# going to a pipe or a StringIO.
1380
'SSH_AUTH_SOCK': None,
1384
'https_proxy': None,
1385
'HTTPS_PROXY': None,
1390
# Nobody cares about ftp_proxy, FTP_PROXY AFAIK. So far at
1391
# least. If you do (care), please update this comment
1395
'BZR_REMOTE_PATH': None,
1398
self.addCleanup(self._restoreEnvironment)
1399
for name, value in new_env.iteritems():
1400
self._captureVar(name, value)
1402
def _captureVar(self, name, newvalue):
1403
"""Set an environment variable, and reset it when finished."""
1404
self.__old_env[name] = osutils.set_or_unset_env(name, newvalue)
1406
def _restore_debug_flags(self):
1407
debug.debug_flags.clear()
1408
debug.debug_flags.update(self._preserved_debug_flags)
1410
def _restoreEnvironment(self):
1411
for name, value in self.__old_env.iteritems():
1412
osutils.set_or_unset_env(name, value)
1414
def _restoreHooks(self):
1415
for klass, (name, hooks) in self._preserved_hooks.items():
1416
setattr(klass, name, hooks)
1418
def knownFailure(self, reason):
1419
"""This test has failed for some known reason."""
1420
raise KnownFailure(reason)
1422
def _do_skip(self, result, reason):
1423
addSkip = getattr(result, 'addSkip', None)
1424
if not callable(addSkip):
1425
result.addError(self, sys.exc_info())
1427
addSkip(self, reason)
1429
def run(self, result=None):
1430
if result is None: result = self.defaultTestResult()
1431
for feature in getattr(self, '_test_needs_features', []):
1432
if not feature.available():
1433
result.startTest(self)
1434
if getattr(result, 'addNotSupported', None):
1435
result.addNotSupported(self, feature)
1437
result.addSuccess(self)
1438
result.stopTest(self)
1442
result.startTest(self)
1443
absent_attr = object()
1445
method_name = getattr(self, '_testMethodName', absent_attr)
1446
if method_name is absent_attr:
1448
method_name = getattr(self, '_TestCase__testMethodName')
1449
testMethod = getattr(self, method_name)
1453
if not self._bzr_test_setUp_run:
1455
"test setUp did not invoke "
1456
"bzrlib.tests.TestCase's setUp")
1457
except KeyboardInterrupt:
1460
except TestSkipped, e:
1461
self._do_skip(result, e.args[0])
1465
result.addError(self, sys.exc_info())
1473
except self.failureException:
1474
result.addFailure(self, sys.exc_info())
1475
except TestSkipped, e:
1477
reason = "No reason given."
1480
self._do_skip(result, reason)
1481
except KeyboardInterrupt:
1485
result.addError(self, sys.exc_info())
1489
if not self._bzr_test_tearDown_run:
1491
"test tearDown did not invoke "
1492
"bzrlib.tests.TestCase's tearDown")
1493
except KeyboardInterrupt:
1497
result.addError(self, sys.exc_info())
1500
if ok: result.addSuccess(self)
1502
result.stopTest(self)
1504
except TestNotApplicable:
1505
# Not moved from the result [yet].
1508
except KeyboardInterrupt:
1513
for attr_name in self.attrs_to_keep:
1514
if attr_name in self.__dict__:
1515
saved_attrs[attr_name] = self.__dict__[attr_name]
1516
self.__dict__ = saved_attrs
1520
self._log_contents = ''
1521
self._bzr_test_tearDown_run = True
1522
unittest.TestCase.tearDown(self)
1524
def time(self, callable, *args, **kwargs):
1525
"""Run callable and accrue the time it takes to the benchmark time.
1527
If lsprofiling is enabled (i.e. by --lsprof-time to bzr selftest) then
1528
this will cause lsprofile statistics to be gathered and stored in
1531
if self._benchtime is None:
1535
if not self._gather_lsprof_in_benchmarks:
1536
return callable(*args, **kwargs)
1538
# record this benchmark
1539
ret, stats = bzrlib.lsprof.profile(callable, *args, **kwargs)
1541
self._benchcalls.append(((callable, args, kwargs), stats))
1544
self._benchtime += time.time() - start
1546
def _runCleanups(self):
1547
"""Run registered cleanup functions.
1549
This should only be called from TestCase.tearDown.
1551
# TODO: Perhaps this should keep running cleanups even if
1552
# one of them fails?
1554
# Actually pop the cleanups from the list so tearDown running
1555
# twice is safe (this happens for skipped tests).
1556
while self._cleanups:
1557
cleanup, args, kwargs = self._cleanups.pop()
1558
cleanup(*args, **kwargs)
1560
def log(self, *args):
1563
def _get_log(self, keep_log_file=False):
1564
"""Get the log from bzrlib.trace calls from this test.
1566
:param keep_log_file: When True, if the log is still a file on disk
1567
leave it as a file on disk. When False, if the log is still a file
1568
on disk, the log file is deleted and the log preserved as
1570
:return: A string containing the log.
1572
# flush the log file, to get all content
1574
if bzrlib.trace._trace_file:
1575
bzrlib.trace._trace_file.flush()
1576
if self._log_contents:
1577
# XXX: this can hardly contain the content flushed above --vila
1579
return self._log_contents
1580
if self._log_file_name is not None:
1581
logfile = open(self._log_file_name)
1583
log_contents = logfile.read()
1586
if not keep_log_file:
1587
self._log_contents = log_contents
1589
os.remove(self._log_file_name)
1591
if sys.platform == 'win32' and e.errno == errno.EACCES:
1592
sys.stderr.write(('Unable to delete log file '
1593
' %r\n' % self._log_file_name))
1598
return "DELETED log file to reduce memory footprint"
1600
def requireFeature(self, feature):
1601
"""This test requires a specific feature is available.
1603
:raises UnavailableFeature: When feature is not available.
1605
if not feature.available():
1606
raise UnavailableFeature(feature)
1608
def _run_bzr_autosplit(self, args, retcode, encoding, stdin,
1610
"""Run bazaar command line, splitting up a string command line."""
1611
if isinstance(args, basestring):
1612
# shlex don't understand unicode strings,
1613
# so args should be plain string (bialix 20070906)
1614
args = list(shlex.split(str(args)))
1615
return self._run_bzr_core(args, retcode=retcode,
1616
encoding=encoding, stdin=stdin, working_dir=working_dir,
1619
def _run_bzr_core(self, args, retcode, encoding, stdin,
1621
# Clear chk_map page cache, because the contents are likely to mask
1623
chk_map.clear_cache()
1624
if encoding is None:
1625
encoding = osutils.get_user_encoding()
1626
stdout = StringIOWrapper()
1627
stderr = StringIOWrapper()
1628
stdout.encoding = encoding
1629
stderr.encoding = encoding
1631
self.log('run bzr: %r', args)
1632
# FIXME: don't call into logging here
1633
handler = logging.StreamHandler(stderr)
1634
handler.setLevel(logging.INFO)
1635
logger = logging.getLogger('')
1636
logger.addHandler(handler)
1637
old_ui_factory = ui.ui_factory
1638
ui.ui_factory = TestUIFactory(stdin=stdin, stdout=stdout, stderr=stderr)
1641
if working_dir is not None:
1642
cwd = osutils.getcwd()
1643
os.chdir(working_dir)
1646
result = self.apply_redirected(ui.ui_factory.stdin,
1648
bzrlib.commands.run_bzr_catch_user_errors,
1651
logger.removeHandler(handler)
1652
ui.ui_factory = old_ui_factory
1656
out = stdout.getvalue()
1657
err = stderr.getvalue()
1659
self.log('output:\n%r', out)
1661
self.log('errors:\n%r', err)
1662
if retcode is not None:
1663
self.assertEquals(retcode, result,
1664
message='Unexpected return code')
1667
def run_bzr(self, args, retcode=0, encoding=None, stdin=None,
1668
working_dir=None, error_regexes=[], output_encoding=None):
1669
"""Invoke bzr, as if it were run from the command line.
1671
The argument list should not include the bzr program name - the
1672
first argument is normally the bzr command. Arguments may be
1673
passed in three ways:
1675
1- A list of strings, eg ["commit", "a"]. This is recommended
1676
when the command contains whitespace or metacharacters, or
1677
is built up at run time.
1679
2- A single string, eg "add a". This is the most convenient
1680
for hardcoded commands.
1682
This runs bzr through the interface that catches and reports
1683
errors, and with logging set to something approximating the
1684
default, so that error reporting can be checked.
1686
This should be the main method for tests that want to exercise the
1687
overall behavior of the bzr application (rather than a unit test
1688
or a functional test of the library.)
1690
This sends the stdout/stderr results into the test's log,
1691
where it may be useful for debugging. See also run_captured.
1693
:keyword stdin: A string to be used as stdin for the command.
1694
:keyword retcode: The status code the command should return;
1696
:keyword working_dir: The directory to run the command in
1697
:keyword error_regexes: A list of expected error messages. If
1698
specified they must be seen in the error output of the command.
1700
out, err = self._run_bzr_autosplit(
1705
working_dir=working_dir,
1707
self.assertIsInstance(error_regexes, (list, tuple))
1708
for regex in error_regexes:
1709
self.assertContainsRe(err, regex)
1712
def run_bzr_error(self, error_regexes, *args, **kwargs):
1713
"""Run bzr, and check that stderr contains the supplied regexes
1715
:param error_regexes: Sequence of regular expressions which
1716
must each be found in the error output. The relative ordering
1718
:param args: command-line arguments for bzr
1719
:param kwargs: Keyword arguments which are interpreted by run_bzr
1720
This function changes the default value of retcode to be 3,
1721
since in most cases this is run when you expect bzr to fail.
1723
:return: (out, err) The actual output of running the command (in case
1724
you want to do more inspection)
1728
# Make sure that commit is failing because there is nothing to do
1729
self.run_bzr_error(['no changes to commit'],
1730
['commit', '-m', 'my commit comment'])
1731
# Make sure --strict is handling an unknown file, rather than
1732
# giving us the 'nothing to do' error
1733
self.build_tree(['unknown'])
1734
self.run_bzr_error(['Commit refused because there are unknown files'],
1735
['commit', --strict', '-m', 'my commit comment'])
1737
kwargs.setdefault('retcode', 3)
1738
kwargs['error_regexes'] = error_regexes
1739
out, err = self.run_bzr(*args, **kwargs)
1742
def run_bzr_subprocess(self, *args, **kwargs):
1743
"""Run bzr in a subprocess for testing.
1745
This starts a new Python interpreter and runs bzr in there.
1746
This should only be used for tests that have a justifiable need for
1747
this isolation: e.g. they are testing startup time, or signal
1748
handling, or early startup code, etc. Subprocess code can't be
1749
profiled or debugged so easily.
1751
:keyword retcode: The status code that is expected. Defaults to 0. If
1752
None is supplied, the status code is not checked.
1753
:keyword env_changes: A dictionary which lists changes to environment
1754
variables. A value of None will unset the env variable.
1755
The values must be strings. The change will only occur in the
1756
child, so you don't need to fix the environment after running.
1757
:keyword universal_newlines: Convert CRLF => LF
1758
:keyword allow_plugins: By default the subprocess is run with
1759
--no-plugins to ensure test reproducibility. Also, it is possible
1760
for system-wide plugins to create unexpected output on stderr,
1761
which can cause unnecessary test failures.
1763
env_changes = kwargs.get('env_changes', {})
1764
working_dir = kwargs.get('working_dir', None)
1765
allow_plugins = kwargs.get('allow_plugins', False)
1767
if isinstance(args[0], list):
1769
elif isinstance(args[0], basestring):
1770
args = list(shlex.split(args[0]))
1772
raise ValueError("passing varargs to run_bzr_subprocess")
1773
process = self.start_bzr_subprocess(args, env_changes=env_changes,
1774
working_dir=working_dir,
1775
allow_plugins=allow_plugins)
1776
# We distinguish between retcode=None and retcode not passed.
1777
supplied_retcode = kwargs.get('retcode', 0)
1778
return self.finish_bzr_subprocess(process, retcode=supplied_retcode,
1779
universal_newlines=kwargs.get('universal_newlines', False),
1782
def start_bzr_subprocess(self, process_args, env_changes=None,
1783
skip_if_plan_to_signal=False,
1785
allow_plugins=False):
1786
"""Start bzr in a subprocess for testing.
1788
This starts a new Python interpreter and runs bzr in there.
1789
This should only be used for tests that have a justifiable need for
1790
this isolation: e.g. they are testing startup time, or signal
1791
handling, or early startup code, etc. Subprocess code can't be
1792
profiled or debugged so easily.
1794
:param process_args: a list of arguments to pass to the bzr executable,
1795
for example ``['--version']``.
1796
:param env_changes: A dictionary which lists changes to environment
1797
variables. A value of None will unset the env variable.
1798
The values must be strings. The change will only occur in the
1799
child, so you don't need to fix the environment after running.
1800
:param skip_if_plan_to_signal: raise TestSkipped when true and os.kill
1802
:param allow_plugins: If False (default) pass --no-plugins to bzr.
1804
:returns: Popen object for the started process.
1806
if skip_if_plan_to_signal:
1807
if not getattr(os, 'kill', None):
1808
raise TestSkipped("os.kill not available.")
1810
if env_changes is None:
1814
def cleanup_environment():
1815
for env_var, value in env_changes.iteritems():
1816
old_env[env_var] = osutils.set_or_unset_env(env_var, value)
1818
def restore_environment():
1819
for env_var, value in old_env.iteritems():
1820
osutils.set_or_unset_env(env_var, value)
1822
bzr_path = self.get_bzr_path()
1825
if working_dir is not None:
1826
cwd = osutils.getcwd()
1827
os.chdir(working_dir)
1830
# win32 subprocess doesn't support preexec_fn
1831
# so we will avoid using it on all platforms, just to
1832
# make sure the code path is used, and we don't break on win32
1833
cleanup_environment()
1834
command = [sys.executable]
1835
# frozen executables don't need the path to bzr
1836
if getattr(sys, "frozen", None) is None:
1837
command.append(bzr_path)
1838
if not allow_plugins:
1839
command.append('--no-plugins')
1840
command.extend(process_args)
1841
process = self._popen(command, stdin=PIPE, stdout=PIPE, stderr=PIPE)
1843
restore_environment()
1849
def _popen(self, *args, **kwargs):
1850
"""Place a call to Popen.
1852
Allows tests to override this method to intercept the calls made to
1853
Popen for introspection.
1855
return Popen(*args, **kwargs)
1857
def get_bzr_path(self):
1858
"""Return the path of the 'bzr' executable for this test suite."""
1859
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
1860
if not os.path.isfile(bzr_path):
1861
# We are probably installed. Assume sys.argv is the right file
1862
bzr_path = sys.argv[0]
1865
def finish_bzr_subprocess(self, process, retcode=0, send_signal=None,
1866
universal_newlines=False, process_args=None):
1867
"""Finish the execution of process.
1869
:param process: the Popen object returned from start_bzr_subprocess.
1870
:param retcode: The status code that is expected. Defaults to 0. If
1871
None is supplied, the status code is not checked.
1872
:param send_signal: an optional signal to send to the process.
1873
:param universal_newlines: Convert CRLF => LF
1874
:returns: (stdout, stderr)
1876
if send_signal is not None:
1877
os.kill(process.pid, send_signal)
1878
out, err = process.communicate()
1880
if universal_newlines:
1881
out = out.replace('\r\n', '\n')
1882
err = err.replace('\r\n', '\n')
1884
if retcode is not None and retcode != process.returncode:
1885
if process_args is None:
1886
process_args = "(unknown args)"
1887
mutter('Output of bzr %s:\n%s', process_args, out)
1888
mutter('Error for bzr %s:\n%s', process_args, err)
1889
self.fail('Command bzr %s failed with retcode %s != %s'
1890
% (process_args, retcode, process.returncode))
125
1893
def check_inventory_shape(self, inv, shape):
127
Compare an inventory to a list of expected names.
1894
"""Compare an inventory to a list of expected names.
129
1896
Fail if they are not precisely equal.
144
1911
self.fail("unexpected paths found in inventory: %r" % extras)
1913
def apply_redirected(self, stdin=None, stdout=None, stderr=None,
1914
a_callable=None, *args, **kwargs):
1915
"""Call callable with redirected std io pipes.
1917
Returns the return code."""
1918
if not callable(a_callable):
1919
raise ValueError("a_callable must be callable.")
1921
stdin = StringIO("")
1923
if getattr(self, "_log_file", None) is not None:
1924
stdout = self._log_file
1928
if getattr(self, "_log_file", None is not None):
1929
stderr = self._log_file
1932
real_stdin = sys.stdin
1933
real_stdout = sys.stdout
1934
real_stderr = sys.stderr
1939
return a_callable(*args, **kwargs)
1941
sys.stdout = real_stdout
1942
sys.stderr = real_stderr
1943
sys.stdin = real_stdin
1945
def reduceLockdirTimeout(self):
1946
"""Reduce the default lock timeout for the duration of the test, so that
1947
if LockContention occurs during a test, it does so quickly.
1949
Tests that expect to provoke LockContention errors should call this.
1951
orig_timeout = bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS
1953
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = orig_timeout
1954
self.addCleanup(resetTimeout)
1955
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = 0
1957
def make_utf8_encoded_stringio(self, encoding_type=None):
1958
"""Return a StringIOWrapper instance, that will encode Unicode
1961
if encoding_type is None:
1962
encoding_type = 'strict'
1964
output_encoding = 'utf-8'
1965
sio = codecs.getwriter(output_encoding)(sio, errors=encoding_type)
1966
sio.encoding = output_encoding
1969
def disable_verb(self, verb):
1970
"""Disable a smart server verb for one test."""
1971
from bzrlib.smart import request
1972
request_handlers = request.request_handlers
1973
orig_method = request_handlers.get(verb)
1974
request_handlers.remove(verb)
1976
request_handlers.register(verb, orig_method)
1977
self.addCleanup(restoreVerb)
1980
class CapturedCall(object):
1981
"""A helper for capturing smart server calls for easy debug analysis."""
1983
def __init__(self, params, prefix_length):
1984
"""Capture the call with params and skip prefix_length stack frames."""
1987
# The last 5 frames are the __init__, the hook frame, and 3 smart
1988
# client frames. Beyond this we could get more clever, but this is good
1990
stack = traceback.extract_stack()[prefix_length:-5]
1991
self.stack = ''.join(traceback.format_list(stack))
1994
return self.call.method
1997
return self.call.method
2003
class TestCaseWithMemoryTransport(TestCase):
2004
"""Common test class for tests that do not need disk resources.
2006
Tests that need disk resources should derive from TestCaseWithTransport.
2008
TestCaseWithMemoryTransport sets the TEST_ROOT variable for all bzr tests.
2010
For TestCaseWithMemoryTransport the test_home_dir is set to the name of
2011
a directory which does not exist. This serves to help ensure test isolation
2012
is preserved. test_dir is set to the TEST_ROOT, as is cwd, because they
2013
must exist. However, TestCaseWithMemoryTransport does not offer local
2014
file defaults for the transport in tests, nor does it obey the command line
2015
override, so tests that accidentally write to the common directory should
2018
:cvar TEST_ROOT: Directory containing all temporary directories, plus
2019
a .bzr directory that stops us ascending higher into the filesystem.
2025
def __init__(self, methodName='runTest'):
2026
# allow test parameterization after test construction and before test
2027
# execution. Variables that the parameterizer sets need to be
2028
# ones that are not set by setUp, or setUp will trash them.
2029
super(TestCaseWithMemoryTransport, self).__init__(methodName)
2030
self.vfs_transport_factory = default_transport
2031
self.transport_server = None
2032
self.transport_readonly_server = None
2033
self.__vfs_server = None
2035
def get_transport(self, relpath=None):
2036
"""Return a writeable transport.
2038
This transport is for the test scratch space relative to
2041
:param relpath: a path relative to the base url.
2043
t = get_transport(self.get_url(relpath))
2044
self.assertFalse(t.is_readonly())
2047
def get_readonly_transport(self, relpath=None):
2048
"""Return a readonly transport for the test scratch space
2050
This can be used to test that operations which should only need
2051
readonly access in fact do not try to write.
2053
:param relpath: a path relative to the base url.
2055
t = get_transport(self.get_readonly_url(relpath))
2056
self.assertTrue(t.is_readonly())
2059
def create_transport_readonly_server(self):
2060
"""Create a transport server from class defined at init.
2062
This is mostly a hook for daughter classes.
2064
return self.transport_readonly_server()
2066
def get_readonly_server(self):
2067
"""Get the server instance for the readonly transport
2069
This is useful for some tests with specific servers to do diagnostics.
2071
if self.__readonly_server is None:
2072
if self.transport_readonly_server is None:
2073
# readonly decorator requested
2074
# bring up the server
2075
self.__readonly_server = ReadonlyServer()
2076
self.__readonly_server.setUp(self.get_vfs_only_server())
2078
self.__readonly_server = self.create_transport_readonly_server()
2079
self.__readonly_server.setUp(self.get_vfs_only_server())
2080
self.addCleanup(self.__readonly_server.tearDown)
2081
return self.__readonly_server
2083
def get_readonly_url(self, relpath=None):
2084
"""Get a URL for the readonly transport.
2086
This will either be backed by '.' or a decorator to the transport
2087
used by self.get_url()
2088
relpath provides for clients to get a path relative to the base url.
2089
These should only be downwards relative, not upwards.
2091
base = self.get_readonly_server().get_url()
2092
return self._adjust_url(base, relpath)
2094
def get_vfs_only_server(self):
2095
"""Get the vfs only read/write server instance.
2097
This is useful for some tests with specific servers that need
2100
For TestCaseWithMemoryTransport this is always a MemoryServer, and there
2101
is no means to override it.
2103
if self.__vfs_server is None:
2104
self.__vfs_server = MemoryServer()
2105
self.__vfs_server.setUp()
2106
self.addCleanup(self.__vfs_server.tearDown)
2107
return self.__vfs_server
2109
def get_server(self):
2110
"""Get the read/write server instance.
2112
This is useful for some tests with specific servers that need
2115
This is built from the self.transport_server factory. If that is None,
2116
then the self.get_vfs_server is returned.
2118
if self.__server is None:
2119
if self.transport_server is None or self.transport_server is self.vfs_transport_factory:
2120
return self.get_vfs_only_server()
2122
# bring up a decorated means of access to the vfs only server.
2123
self.__server = self.transport_server()
2125
self.__server.setUp(self.get_vfs_only_server())
2126
except TypeError, e:
2127
# This should never happen; the try:Except here is to assist
2128
# developers having to update code rather than seeing an
2129
# uninformative TypeError.
2130
raise Exception, "Old server API in use: %s, %s" % (self.__server, e)
2131
self.addCleanup(self.__server.tearDown)
2132
return self.__server
2134
def _adjust_url(self, base, relpath):
2135
"""Get a URL (or maybe a path) for the readwrite transport.
2137
This will either be backed by '.' or to an equivalent non-file based
2139
relpath provides for clients to get a path relative to the base url.
2140
These should only be downwards relative, not upwards.
2142
if relpath is not None and relpath != '.':
2143
if not base.endswith('/'):
2145
# XXX: Really base should be a url; we did after all call
2146
# get_url()! But sometimes it's just a path (from
2147
# LocalAbspathServer), and it'd be wrong to append urlescaped data
2148
# to a non-escaped local path.
2149
if base.startswith('./') or base.startswith('/'):
2152
base += urlutils.escape(relpath)
2155
def get_url(self, relpath=None):
2156
"""Get a URL (or maybe a path) for the readwrite transport.
2158
This will either be backed by '.' or to an equivalent non-file based
2160
relpath provides for clients to get a path relative to the base url.
2161
These should only be downwards relative, not upwards.
2163
base = self.get_server().get_url()
2164
return self._adjust_url(base, relpath)
2166
def get_vfs_only_url(self, relpath=None):
2167
"""Get a URL (or maybe a path for the plain old vfs transport.
2169
This will never be a smart protocol. It always has all the
2170
capabilities of the local filesystem, but it might actually be a
2171
MemoryTransport or some other similar virtual filesystem.
2173
This is the backing transport (if any) of the server returned by
2174
get_url and get_readonly_url.
2176
:param relpath: provides for clients to get a path relative to the base
2177
url. These should only be downwards relative, not upwards.
2180
base = self.get_vfs_only_server().get_url()
2181
return self._adjust_url(base, relpath)
2183
def _create_safety_net(self):
2184
"""Make a fake bzr directory.
2186
This prevents any tests propagating up onto the TEST_ROOT directory's
2189
root = TestCaseWithMemoryTransport.TEST_ROOT
2190
bzrdir.BzrDir.create_standalone_workingtree(root)
2192
def _check_safety_net(self):
2193
"""Check that the safety .bzr directory have not been touched.
2195
_make_test_root have created a .bzr directory to prevent tests from
2196
propagating. This method ensures than a test did not leaked.
2198
root = TestCaseWithMemoryTransport.TEST_ROOT
2199
wt = workingtree.WorkingTree.open(root)
2200
last_rev = wt.last_revision()
2201
if last_rev != 'null:':
2202
# The current test have modified the /bzr directory, we need to
2203
# recreate a new one or all the followng tests will fail.
2204
# If you need to inspect its content uncomment the following line
2205
# import pdb; pdb.set_trace()
2206
_rmtree_temp_dir(root + '/.bzr')
2207
self._create_safety_net()
2208
raise AssertionError('%s/.bzr should not be modified' % root)
2210
def _make_test_root(self):
2211
if TestCaseWithMemoryTransport.TEST_ROOT is None:
2212
root = osutils.mkdtemp(prefix='testbzr-', suffix='.tmp')
2213
TestCaseWithMemoryTransport.TEST_ROOT = root
2215
self._create_safety_net()
2217
# The same directory is used by all tests, and we're not
2218
# specifically told when all tests are finished. This will do.
2219
atexit.register(_rmtree_temp_dir, root)
2221
self.addCleanup(self._check_safety_net)
2223
def makeAndChdirToTestDir(self):
2224
"""Create a temporary directories for this one test.
2226
This must set self.test_home_dir and self.test_dir and chdir to
2229
For TestCaseWithMemoryTransport we chdir to the TEST_ROOT for this test.
2231
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2232
self.test_dir = TestCaseWithMemoryTransport.TEST_ROOT
2233
self.test_home_dir = self.test_dir + "/MemoryTransportMissingHomeDir"
2235
def make_branch(self, relpath, format=None):
2236
"""Create a branch on the transport at relpath."""
2237
repo = self.make_repository(relpath, format=format)
2238
return repo.bzrdir.create_branch()
2240
def make_bzrdir(self, relpath, format=None):
2242
# might be a relative or absolute path
2243
maybe_a_url = self.get_url(relpath)
2244
segments = maybe_a_url.rsplit('/', 1)
2245
t = get_transport(maybe_a_url)
2246
if len(segments) > 1 and segments[-1] not in ('', '.'):
2250
if isinstance(format, basestring):
2251
format = bzrdir.format_registry.make_bzrdir(format)
2252
return format.initialize_on_transport(t)
2253
except errors.UninitializableFormat:
2254
raise TestSkipped("Format %s is not initializable." % format)
2256
def make_repository(self, relpath, shared=False, format=None):
2257
"""Create a repository on our default transport at relpath.
2259
Note that relpath must be a relative path, not a full url.
2261
# FIXME: If you create a remoterepository this returns the underlying
2262
# real format, which is incorrect. Actually we should make sure that
2263
# RemoteBzrDir returns a RemoteRepository.
2264
# maybe mbp 20070410
2265
made_control = self.make_bzrdir(relpath, format=format)
2266
return made_control.create_repository(shared=shared)
2268
def make_smart_server(self, path):
2269
smart_server = server.SmartTCPServer_for_testing()
2270
smart_server.setUp(self.get_server())
2271
remote_transport = get_transport(smart_server.get_url()).clone(path)
2272
self.addCleanup(smart_server.tearDown)
2273
return remote_transport
2275
def make_branch_and_memory_tree(self, relpath, format=None):
2276
"""Create a branch on the default transport and a MemoryTree for it."""
2277
b = self.make_branch(relpath, format=format)
2278
return memorytree.MemoryTree.create_on_branch(b)
2280
def make_branch_builder(self, relpath, format=None):
2281
branch = self.make_branch(relpath, format=format)
2282
return branchbuilder.BranchBuilder(branch=branch)
2284
def overrideEnvironmentForTesting(self):
2285
os.environ['HOME'] = self.test_home_dir
2286
os.environ['BZR_HOME'] = self.test_home_dir
2289
super(TestCaseWithMemoryTransport, self).setUp()
2290
self._make_test_root()
2291
_currentdir = os.getcwdu()
2292
def _leaveDirectory():
2293
os.chdir(_currentdir)
2294
self.addCleanup(_leaveDirectory)
2295
self.makeAndChdirToTestDir()
2296
self.overrideEnvironmentForTesting()
2297
self.__readonly_server = None
2298
self.__server = None
2299
self.reduceLockdirTimeout()
2301
def setup_smart_server_with_call_log(self):
2302
"""Sets up a smart server as the transport server with a call log."""
2303
self.transport_server = server.SmartTCPServer_for_testing
2304
self.hpss_calls = []
2306
# Skip the current stack down to the caller of
2307
# setup_smart_server_with_call_log
2308
prefix_length = len(traceback.extract_stack()) - 2
2309
def capture_hpss_call(params):
2310
self.hpss_calls.append(
2311
CapturedCall(params, prefix_length))
2312
client._SmartClient.hooks.install_named_hook(
2313
'call', capture_hpss_call, None)
2315
def reset_smart_call_log(self):
2316
self.hpss_calls = []
2319
class TestCaseInTempDir(TestCaseWithMemoryTransport):
2320
"""Derived class that runs a test within a temporary directory.
2322
This is useful for tests that need to create a branch, etc.
2324
The directory is created in a slightly complex way: for each
2325
Python invocation, a new temporary top-level directory is created.
2326
All test cases create their own directory within that. If the
2327
tests complete successfully, the directory is removed.
2329
:ivar test_base_dir: The path of the top-level directory for this
2330
test, which contains a home directory and a work directory.
2332
:ivar test_home_dir: An initially empty directory under test_base_dir
2333
which is used as $HOME for this test.
2335
:ivar test_dir: A directory under test_base_dir used as the current
2336
directory when the test proper is run.
2339
OVERRIDE_PYTHON = 'python'
147
2341
def check_file_contents(self, filename, expect):
148
2342
self.log("check contents of file %s" % filename)
149
2343
contents = file(filename, 'r').read()
150
2344
if contents != expect:
151
self.log("expected: %r" % expected)
2345
self.log("expected: %r" % expect)
152
2346
self.log("actually: %r" % contents)
153
self.fail("contents of %s not as expected")
157
class InTempDir(TestBase):
158
"""Base class for tests run in a temporary branch."""
161
self.test_dir = os.path.join(self.TEST_ROOT, self.__class__.__name__)
2347
self.fail("contents of %s not as expected" % filename)
2349
def _getTestDirPrefix(self):
2350
# create a directory within the top level test directory
2351
if sys.platform in ('win32', 'cygwin'):
2352
name_prefix = re.sub('[<>*=+",:;_/\\-]', '_', self.id())
2353
# windows is likely to have path-length limits so use a short name
2354
name_prefix = name_prefix[-30:]
2356
name_prefix = re.sub('[/]', '_', self.id())
2359
def makeAndChdirToTestDir(self):
2360
"""See TestCaseWithMemoryTransport.makeAndChdirToTestDir().
2362
For TestCaseInTempDir we create a temporary directory based on the test
2363
name and then create two subdirs - test and home under it.
2365
name_prefix = osutils.pathjoin(TestCaseWithMemoryTransport.TEST_ROOT,
2366
self._getTestDirPrefix())
2368
for i in range(100):
2369
if os.path.exists(name):
2370
name = name_prefix + '_' + str(i)
2374
# now create test and home directories within this dir
2375
self.test_base_dir = name
2376
self.test_home_dir = self.test_base_dir + '/home'
2377
os.mkdir(self.test_home_dir)
2378
self.test_dir = self.test_base_dir + '/work'
162
2379
os.mkdir(self.test_dir)
163
2380
os.chdir(self.test_dir)
167
os.chdir(self.TEST_ROOT)
173
class _MyResult(TestResult):
177
No special behaviour for now.
179
def __init__(self, out):
181
TestResult.__init__(self)
2381
# put name of test inside
2382
f = file(self.test_base_dir + '/name', 'w')
2387
self.addCleanup(self.deleteTestDir)
2389
def deleteTestDir(self):
2390
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
2391
_rmtree_temp_dir(self.test_base_dir)
2393
def build_tree(self, shape, line_endings='binary', transport=None):
2394
"""Build a test tree according to a pattern.
2396
shape is a sequence of file specifications. If the final
2397
character is '/', a directory is created.
2399
This assumes that all the elements in the tree being built are new.
2401
This doesn't add anything to a branch.
2403
:type shape: list or tuple.
2404
:param line_endings: Either 'binary' or 'native'
2405
in binary mode, exact contents are written in native mode, the
2406
line endings match the default platform endings.
2407
:param transport: A transport to write to, for building trees on VFS's.
2408
If the transport is readonly or None, "." is opened automatically.
2411
if type(shape) not in (list, tuple):
2412
raise AssertionError("Parameter 'shape' should be "
2413
"a list or a tuple. Got %r instead" % (shape,))
2414
# It's OK to just create them using forward slashes on windows.
2415
if transport is None or transport.is_readonly():
2416
transport = get_transport(".")
2418
self.assertIsInstance(name, basestring)
2420
transport.mkdir(urlutils.escape(name[:-1]))
2422
if line_endings == 'binary':
2424
elif line_endings == 'native':
2427
raise errors.BzrError(
2428
'Invalid line ending request %r' % line_endings)
2429
content = "contents of %s%s" % (name.encode('utf-8'), end)
2430
transport.put_bytes_non_atomic(urlutils.escape(name), content)
2432
def build_tree_contents(self, shape):
2433
build_tree_contents(shape)
2435
def assertInWorkingTree(self, path, root_path='.', tree=None):
2436
"""Assert whether path or paths are in the WorkingTree"""
2438
tree = workingtree.WorkingTree.open(root_path)
2439
if not isinstance(path, basestring):
2441
self.assertInWorkingTree(p, tree=tree)
2443
self.assertIsNot(tree.path2id(path), None,
2444
path+' not in working tree.')
2446
def assertNotInWorkingTree(self, path, root_path='.', tree=None):
2447
"""Assert whether path or paths are not in the WorkingTree"""
2449
tree = workingtree.WorkingTree.open(root_path)
2450
if not isinstance(path, basestring):
2452
self.assertNotInWorkingTree(p,tree=tree)
2454
self.assertIs(tree.path2id(path), None, path+' in working tree.')
2457
class TestCaseWithTransport(TestCaseInTempDir):
2458
"""A test case that provides get_url and get_readonly_url facilities.
2460
These back onto two transport servers, one for readonly access and one for
2463
If no explicit class is provided for readonly access, a
2464
ReadonlyTransportDecorator is used instead which allows the use of non disk
2465
based read write transports.
2467
If an explicit class is provided for readonly access, that server and the
2468
readwrite one must both define get_url() as resolving to os.getcwd().
2471
def get_vfs_only_server(self):
2472
"""See TestCaseWithMemoryTransport.
2474
This is useful for some tests with specific servers that need
2477
if self.__vfs_server is None:
2478
self.__vfs_server = self.vfs_transport_factory()
2479
self.__vfs_server.setUp()
2480
self.addCleanup(self.__vfs_server.tearDown)
2481
return self.__vfs_server
2483
def make_branch_and_tree(self, relpath, format=None):
2484
"""Create a branch on the transport and a tree locally.
2486
If the transport is not a LocalTransport, the Tree can't be created on
2487
the transport. In that case if the vfs_transport_factory is
2488
LocalURLServer the working tree is created in the local
2489
directory backing the transport, and the returned tree's branch and
2490
repository will also be accessed locally. Otherwise a lightweight
2491
checkout is created and returned.
2493
:param format: The BzrDirFormat.
2494
:returns: the WorkingTree.
2496
# TODO: always use the local disk path for the working tree,
2497
# this obviously requires a format that supports branch references
2498
# so check for that by checking bzrdir.BzrDirFormat.get_default_format()
2500
b = self.make_branch(relpath, format=format)
2502
return b.bzrdir.create_workingtree()
2503
except errors.NotLocalUrl:
2504
# We can only make working trees locally at the moment. If the
2505
# transport can't support them, then we keep the non-disk-backed
2506
# branch and create a local checkout.
2507
if self.vfs_transport_factory is LocalURLServer:
2508
# the branch is colocated on disk, we cannot create a checkout.
2509
# hopefully callers will expect this.
2510
local_controldir= bzrdir.BzrDir.open(self.get_vfs_only_url(relpath))
2511
wt = local_controldir.create_workingtree()
2512
if wt.branch._format != b._format:
2514
# Make sure that assigning to wt._branch fixes wt.branch,
2515
# in case the implementation details of workingtree objects
2517
self.assertIs(b, wt.branch)
2520
return b.create_checkout(relpath, lightweight=True)
2522
def assertIsDirectory(self, relpath, transport):
2523
"""Assert that relpath within transport is a directory.
2525
This may not be possible on all transports; in that case it propagates
2526
a TransportNotPossible.
2529
mode = transport.stat(relpath).st_mode
2530
except errors.NoSuchFile:
2531
self.fail("path %s is not a directory; no such file"
2533
if not stat.S_ISDIR(mode):
2534
self.fail("path %s is not a directory; has mode %#o"
2537
def assertTreesEqual(self, left, right):
2538
"""Check that left and right have the same content and properties."""
2539
# we use a tree delta to check for equality of the content, and we
2540
# manually check for equality of other things such as the parents list.
2541
self.assertEqual(left.get_parent_ids(), right.get_parent_ids())
2542
differences = left.changes_from(right)
2543
self.assertFalse(differences.has_changed(),
2544
"Trees %r and %r are different: %r" % (left, right, differences))
2547
super(TestCaseWithTransport, self).setUp()
2548
self.__vfs_server = None
2551
class ChrootedTestCase(TestCaseWithTransport):
2552
"""A support class that provides readonly urls outside the local namespace.
2554
This is done by checking if self.transport_server is a MemoryServer. if it
2555
is then we are chrooted already, if it is not then an HttpServer is used
2558
TODO RBC 20060127: make this an option to TestCaseWithTransport so it can
2559
be used without needed to redo it when a different
2560
subclass is in use ?
2564
super(ChrootedTestCase, self).setUp()
2565
if not self.vfs_transport_factory == MemoryServer:
2566
self.transport_readonly_server = HttpServer
2569
def condition_id_re(pattern):
2570
"""Create a condition filter which performs a re check on a test's id.
2572
:param pattern: A regular expression string.
2573
:return: A callable that returns True if the re matches.
2575
filter_re = osutils.re_compile_checked(pattern, 0,
2577
def condition(test):
2579
return filter_re.search(test_id)
2583
def condition_isinstance(klass_or_klass_list):
2584
"""Create a condition filter which returns isinstance(param, klass).
2586
:return: A callable which when called with one parameter obj return the
2587
result of isinstance(obj, klass_or_klass_list).
2590
return isinstance(obj, klass_or_klass_list)
2594
def condition_id_in_list(id_list):
2595
"""Create a condition filter which verify that test's id in a list.
2597
:param id_list: A TestIdList object.
2598
:return: A callable that returns True if the test's id appears in the list.
2600
def condition(test):
2601
return id_list.includes(test.id())
2605
def condition_id_startswith(starts):
2606
"""Create a condition filter verifying that test's id starts with a string.
2608
:param starts: A list of string.
2609
:return: A callable that returns True if the test's id starts with one of
2612
def condition(test):
2613
for start in starts:
2614
if test.id().startswith(start):
2620
def exclude_tests_by_condition(suite, condition):
2621
"""Create a test suite which excludes some tests from suite.
2623
:param suite: The suite to get tests from.
2624
:param condition: A callable whose result evaluates True when called with a
2625
test case which should be excluded from the result.
2626
:return: A suite which contains the tests found in suite that fail
2630
for test in iter_suite_tests(suite):
2631
if not condition(test):
2633
return TestUtil.TestSuite(result)
2636
def filter_suite_by_condition(suite, condition):
2637
"""Create a test suite by filtering another one.
2639
:param suite: The source suite.
2640
:param condition: A callable whose result evaluates True when called with a
2641
test case which should be included in the result.
2642
:return: A suite which contains the tests found in suite that pass
2646
for test in iter_suite_tests(suite):
2649
return TestUtil.TestSuite(result)
2652
def filter_suite_by_re(suite, pattern):
2653
"""Create a test suite by filtering another one.
2655
:param suite: the source suite
2656
:param pattern: pattern that names must match
2657
:returns: the newly created suite
2659
condition = condition_id_re(pattern)
2660
result_suite = filter_suite_by_condition(suite, condition)
2664
def filter_suite_by_id_list(suite, test_id_list):
2665
"""Create a test suite by filtering another one.
2667
:param suite: The source suite.
2668
:param test_id_list: A list of the test ids to keep as strings.
2669
:returns: the newly created suite
2671
condition = condition_id_in_list(test_id_list)
2672
result_suite = filter_suite_by_condition(suite, condition)
2676
def filter_suite_by_id_startswith(suite, start):
2677
"""Create a test suite by filtering another one.
2679
:param suite: The source suite.
2680
:param start: A list of string the test id must start with one of.
2681
:returns: the newly created suite
2683
condition = condition_id_startswith(start)
2684
result_suite = filter_suite_by_condition(suite, condition)
2688
def exclude_tests_by_re(suite, pattern):
2689
"""Create a test suite which excludes some tests from suite.
2691
:param suite: The suite to get tests from.
2692
:param pattern: A regular expression string. Test ids that match this
2693
pattern will be excluded from the result.
2694
:return: A TestSuite that contains all the tests from suite without the
2695
tests that matched pattern. The order of tests is the same as it was in
2698
return exclude_tests_by_condition(suite, condition_id_re(pattern))
2701
def preserve_input(something):
2702
"""A helper for performing test suite transformation chains.
2704
:param something: Anything you want to preserve.
2710
def randomize_suite(suite):
2711
"""Return a new TestSuite with suite's tests in random order.
2713
The tests in the input suite are flattened into a single suite in order to
2714
accomplish this. Any nested TestSuites are removed to provide global
2717
tests = list(iter_suite_tests(suite))
2718
random.shuffle(tests)
2719
return TestUtil.TestSuite(tests)
2722
def split_suite_by_condition(suite, condition):
2723
"""Split a test suite into two by a condition.
2725
:param suite: The suite to split.
2726
:param condition: The condition to match on. Tests that match this
2727
condition are returned in the first test suite, ones that do not match
2728
are in the second suite.
2729
:return: A tuple of two test suites, where the first contains tests from
2730
suite matching the condition, and the second contains the remainder
2731
from suite. The order within each output suite is the same as it was in
2736
for test in iter_suite_tests(suite):
2738
matched.append(test)
2740
did_not_match.append(test)
2741
return TestUtil.TestSuite(matched), TestUtil.TestSuite(did_not_match)
2744
def split_suite_by_re(suite, pattern):
2745
"""Split a test suite into two by a regular expression.
2747
:param suite: The suite to split.
2748
:param pattern: A regular expression string. Test ids that match this
2749
pattern will be in the first test suite returned, and the others in the
2750
second test suite returned.
2751
:return: A tuple of two test suites, where the first contains tests from
2752
suite matching pattern, and the second contains the remainder from
2753
suite. The order within each output suite is the same as it was in
2756
return split_suite_by_condition(suite, condition_id_re(pattern))
2759
def run_suite(suite, name='test', verbose=False, pattern=".*",
2760
stop_on_failure=False,
2761
transport=None, lsprof_timed=None, bench_history=None,
2762
matching_tests_first=None,
2765
exclude_pattern=None,
2768
suite_decorators=None,
2770
"""Run a test suite for bzr selftest.
2772
:param runner_class: The class of runner to use. Must support the
2773
constructor arguments passed by run_suite which are more than standard
2775
:return: A boolean indicating success.
2777
TestCase._gather_lsprof_in_benchmarks = lsprof_timed
2782
if runner_class is None:
2783
runner_class = TextTestRunner
2786
runner = runner_class(stream=stream,
2788
verbosity=verbosity,
2789
bench_history=bench_history,
2790
list_only=list_only,
2793
runner.stop_on_failure=stop_on_failure
2794
# built in decorator factories:
2796
random_order(random_seed, runner),
2797
exclude_tests(exclude_pattern),
2799
if matching_tests_first:
2800
decorators.append(tests_first(pattern))
2802
decorators.append(filter_tests(pattern))
2803
if suite_decorators:
2804
decorators.extend(suite_decorators)
2805
# tell the result object how many tests will be running: (except if
2806
# --parallel=fork is being used. Robert said he will provide a better
2807
# progress design later -- vila 20090817)
2808
if fork_decorator not in decorators:
2809
decorators.append(CountingDecorator)
2810
for decorator in decorators:
2811
suite = decorator(suite)
2812
result = runner.run(suite)
2817
return result.wasStrictlySuccessful()
2819
return result.wasSuccessful()
2822
# A registry where get() returns a suite decorator.
2823
parallel_registry = registry.Registry()
2826
def fork_decorator(suite):
2827
concurrency = osutils.local_concurrency()
2828
if concurrency == 1:
2830
from testtools import ConcurrentTestSuite
2831
return ConcurrentTestSuite(suite, fork_for_tests)
2832
parallel_registry.register('fork', fork_decorator)
2835
def subprocess_decorator(suite):
2836
concurrency = osutils.local_concurrency()
2837
if concurrency == 1:
2839
from testtools import ConcurrentTestSuite
2840
return ConcurrentTestSuite(suite, reinvoke_for_tests)
2841
parallel_registry.register('subprocess', subprocess_decorator)
2844
def exclude_tests(exclude_pattern):
2845
"""Return a test suite decorator that excludes tests."""
2846
if exclude_pattern is None:
2847
return identity_decorator
2848
def decorator(suite):
2849
return ExcludeDecorator(suite, exclude_pattern)
2853
def filter_tests(pattern):
2855
return identity_decorator
2856
def decorator(suite):
2857
return FilterTestsDecorator(suite, pattern)
2861
def random_order(random_seed, runner):
2862
"""Return a test suite decorator factory for randomising tests order.
2864
:param random_seed: now, a string which casts to a long, or a long.
2865
:param runner: A test runner with a stream attribute to report on.
2867
if random_seed is None:
2868
return identity_decorator
2869
def decorator(suite):
2870
return RandomDecorator(suite, random_seed, runner.stream)
2874
def tests_first(pattern):
2876
return identity_decorator
2877
def decorator(suite):
2878
return TestFirstDecorator(suite, pattern)
2882
def identity_decorator(suite):
2887
class TestDecorator(TestSuite):
2888
"""A decorator for TestCase/TestSuite objects.
2890
Usually, subclasses should override __iter__(used when flattening test
2891
suites), which we do to filter, reorder, parallelise and so on, run() and
2895
def __init__(self, suite):
2896
TestSuite.__init__(self)
2899
def countTestCases(self):
2902
cases += test.countTestCases()
2909
def run(self, result):
2910
# Use iteration on self, not self._tests, to allow subclasses to hook
2913
if result.shouldStop:
2919
class CountingDecorator(TestDecorator):
2920
"""A decorator which calls result.progress(self.countTestCases)."""
2922
def run(self, result):
2923
progress_method = getattr(result, 'progress', None)
2924
if callable(progress_method):
2925
progress_method(self.countTestCases(), SUBUNIT_SEEK_SET)
2926
return super(CountingDecorator, self).run(result)
2929
class ExcludeDecorator(TestDecorator):
2930
"""A decorator which excludes test matching an exclude pattern."""
2932
def __init__(self, suite, exclude_pattern):
2933
TestDecorator.__init__(self, suite)
2934
self.exclude_pattern = exclude_pattern
2935
self.excluded = False
2939
return iter(self._tests)
2940
self.excluded = True
2941
suite = exclude_tests_by_re(self, self.exclude_pattern)
2943
self.addTests(suite)
2944
return iter(self._tests)
2947
class FilterTestsDecorator(TestDecorator):
2948
"""A decorator which filters tests to those matching a pattern."""
2950
def __init__(self, suite, pattern):
2951
TestDecorator.__init__(self, suite)
2952
self.pattern = pattern
2953
self.filtered = False
2957
return iter(self._tests)
2958
self.filtered = True
2959
suite = filter_suite_by_re(self, self.pattern)
2961
self.addTests(suite)
2962
return iter(self._tests)
2965
class RandomDecorator(TestDecorator):
2966
"""A decorator which randomises the order of its tests."""
2968
def __init__(self, suite, random_seed, stream):
2969
TestDecorator.__init__(self, suite)
2970
self.random_seed = random_seed
2971
self.randomised = False
2972
self.stream = stream
2976
return iter(self._tests)
2977
self.randomised = True
2978
self.stream.writeln("Randomizing test order using seed %s\n" %
2979
(self.actual_seed()))
2980
# Initialise the random number generator.
2981
random.seed(self.actual_seed())
2982
suite = randomize_suite(self)
2984
self.addTests(suite)
2985
return iter(self._tests)
2987
def actual_seed(self):
2988
if self.random_seed == "now":
2989
# We convert the seed to a long to make it reuseable across
2990
# invocations (because the user can reenter it).
2991
self.random_seed = long(time.time())
2993
# Convert the seed to a long if we can
2995
self.random_seed = long(self.random_seed)
2998
return self.random_seed
3001
class TestFirstDecorator(TestDecorator):
3002
"""A decorator which moves named tests to the front."""
3004
def __init__(self, suite, pattern):
3005
TestDecorator.__init__(self, suite)
3006
self.pattern = pattern
3007
self.filtered = False
3011
return iter(self._tests)
3012
self.filtered = True
3013
suites = split_suite_by_re(self, self.pattern)
3015
self.addTests(suites)
3016
return iter(self._tests)
3019
def partition_tests(suite, count):
3020
"""Partition suite into count lists of tests."""
3022
tests = list(iter_suite_tests(suite))
3023
tests_per_process = int(math.ceil(float(len(tests)) / count))
3024
for block in range(count):
3025
low_test = block * tests_per_process
3026
high_test = low_test + tests_per_process
3027
process_tests = tests[low_test:high_test]
3028
result.append(process_tests)
3032
def fork_for_tests(suite):
3033
"""Take suite and start up one runner per CPU by forking()
3035
:return: An iterable of TestCase-like objects which can each have
3036
run(result) called on them to feed tests to result.
3038
concurrency = osutils.local_concurrency()
3040
from subunit import TestProtocolClient, ProtocolTestCase
3042
from subunit.test_results import AutoTimingTestResultDecorator
3044
AutoTimingTestResultDecorator = lambda x:x
3045
class TestInOtherProcess(ProtocolTestCase):
3046
# Should be in subunit, I think. RBC.
3047
def __init__(self, stream, pid):
3048
ProtocolTestCase.__init__(self, stream)
3051
def run(self, result):
3053
ProtocolTestCase.run(self, result)
3055
os.waitpid(self.pid, os.WNOHANG)
3057
test_blocks = partition_tests(suite, concurrency)
3058
for process_tests in test_blocks:
3059
process_suite = TestSuite()
3060
process_suite.addTests(process_tests)
3061
c2pread, c2pwrite = os.pipe()
3066
# Leave stderr and stdout open so we can see test noise
3067
# Close stdin so that the child goes away if it decides to
3068
# read from stdin (otherwise its a roulette to see what
3069
# child actually gets keystrokes for pdb etc).
3072
stream = os.fdopen(c2pwrite, 'wb', 1)
3073
subunit_result = AutoTimingTestResultDecorator(
3074
TestProtocolClient(stream))
3075
process_suite.run(subunit_result)
3080
stream = os.fdopen(c2pread, 'rb', 1)
3081
test = TestInOtherProcess(stream, pid)
3086
def reinvoke_for_tests(suite):
3087
"""Take suite and start up one runner per CPU using subprocess().
3089
:return: An iterable of TestCase-like objects which can each have
3090
run(result) called on them to feed tests to result.
3092
concurrency = osutils.local_concurrency()
3094
from subunit import ProtocolTestCase
3095
class TestInSubprocess(ProtocolTestCase):
3096
def __init__(self, process, name):
3097
ProtocolTestCase.__init__(self, process.stdout)
3098
self.process = process
3099
self.process.stdin.close()
3102
def run(self, result):
3104
ProtocolTestCase.run(self, result)
3107
os.unlink(self.name)
3108
# print "pid %d finished" % finished_process
3109
test_blocks = partition_tests(suite, concurrency)
3110
for process_tests in test_blocks:
3111
# ugly; currently reimplement rather than reuses TestCase methods.
3112
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
3113
if not os.path.isfile(bzr_path):
3114
# We are probably installed. Assume sys.argv is the right file
3115
bzr_path = sys.argv[0]
3116
fd, test_list_file_name = tempfile.mkstemp()
3117
test_list_file = os.fdopen(fd, 'wb', 1)
3118
for test in process_tests:
3119
test_list_file.write(test.id() + '\n')
3120
test_list_file.close()
3122
argv = [bzr_path, 'selftest', '--load-list', test_list_file_name,
3124
if '--no-plugins' in sys.argv:
3125
argv.append('--no-plugins')
3126
# stderr=STDOUT would be ideal, but until we prevent noise on
3127
# stderr it can interrupt the subunit protocol.
3128
process = Popen(argv, stdin=PIPE, stdout=PIPE, stderr=PIPE,
3130
test = TestInSubprocess(process, test_list_file_name)
3133
os.unlink(test_list_file_name)
3138
class BZRTransformingResult(unittest.TestResult):
3140
def __init__(self, target):
3141
unittest.TestResult.__init__(self)
3142
self.result = target
183
3144
def startTest(self, test):
184
# TODO: Maybe show test.shortDescription somewhere?
185
print >>self.out, '%-60.60s' % test.id(),
187
TestResult.startTest(self, test)
3145
self.result.startTest(test)
189
3147
def stopTest(self, test):
191
TestResult.stopTest(self, test)
3148
self.result.stopTest(test)
194
3150
def addError(self, test, err):
195
print >>self.out, 'ERROR'
196
TestResult.addError(self, test, err)
197
_show_test_failure('error', test, err, self.out)
3151
feature = self._error_looks_like('UnavailableFeature: ', err)
3152
if feature is not None:
3153
self.result.addNotSupported(test, feature)
3155
self.result.addError(test, err)
199
3157
def addFailure(self, test, err):
200
print >>self.out, 'FAILURE'
201
TestResult.addFailure(self, test, err)
202
_show_test_failure('failure', test, err, self.out)
3158
known = self._error_looks_like('KnownFailure: ', err)
3159
if known is not None:
3160
self.result._addKnownFailure(test, [KnownFailure,
3161
KnownFailure(known), None])
3163
self.result.addFailure(test, err)
3165
def addSkip(self, test, reason):
3166
self.result.addSkip(test, reason)
204
3168
def addSuccess(self, test):
205
print >>self.out, 'OK'
206
TestResult.addSuccess(self, test)
211
from unittest import TestLoader, TestSuite
212
import bzrlib, bzrlib.store, bzrlib.inventory, bzrlib.branch, bzrlib.osutils, bzrlib.commands
214
import bzrlib.selftest.whitebox
215
import bzrlib.selftest.blackbox
216
import bzrlib.selftest.versioning
217
from doctest import DocTestSuite
223
TestBase.BZRPATH = os.path.join(os.path.realpath(os.path.dirname(bzrlib.__path__[0])), 'bzr')
224
print '%-30s %s' % ('bzr binary', TestBase.BZRPATH)
233
for m in bzrlib.selftest.whitebox, \
234
bzrlib.selftest.versioning:
235
suite.addTest(tl.loadTestsFromModule(m))
237
for m in bzrlib.store, bzrlib.inventory, bzrlib.branch, bzrlib.osutils, \
239
suite.addTest(DocTestSuite(m))
241
suite.addTest(bzrlib.selftest.blackbox.suite())
243
# save stdout & stderr so there's no leakage from code-under-test
244
real_stdout = sys.stdout
245
real_stderr = sys.stderr
246
sys.stdout = sys.stderr = TestBase.TEST_LOG
3169
self.result.addSuccess(test)
3171
def _error_looks_like(self, prefix, err):
3172
"""Deserialize exception and returns the stringify value."""
3176
if isinstance(exc, subunit.RemoteException):
3177
# stringify the exception gives access to the remote traceback
3178
# We search the last line for 'prefix'
3179
lines = str(exc).split('\n')
3180
while lines and not lines[-1]:
3183
if lines[-1].startswith(prefix):
3184
value = lines[-1][len(prefix):]
3188
# Controlled by "bzr selftest -E=..." option
3189
# Currently supported:
3190
# -Eallow_debug Will no longer clear debug.debug_flags() so it
3191
# preserves any flags supplied at the command line.
3192
# -Edisable_lock_checks Turns errors in mismatched locks into simple prints
3193
# rather than failing tests. And no longer raise
3194
# LockContention when fctnl locks are not being used
3195
# with proper exclusion rules.
3196
selftest_debug_flags = set()
3199
def selftest(verbose=False, pattern=".*", stop_on_failure=True,
3201
test_suite_factory=None,
3204
matching_tests_first=None,
3207
exclude_pattern=None,
3213
suite_decorators=None,
3216
"""Run the whole test suite under the enhanced runner"""
3217
# XXX: Very ugly way to do this...
3218
# Disable warning about old formats because we don't want it to disturb
3219
# any blackbox tests.
3220
from bzrlib import repository
3221
repository._deprecation_warning_done = True
3223
global default_transport
3224
if transport is None:
3225
transport = default_transport
3226
old_transport = default_transport
3227
default_transport = transport
3228
global selftest_debug_flags
3229
old_debug_flags = selftest_debug_flags
3230
if debug_flags is not None:
3231
selftest_debug_flags = set(debug_flags)
248
result = _MyResult(real_stdout)
3233
if load_list is None:
3236
keep_only = load_test_id_list(load_list)
3238
starting_with = [test_prefix_alias_registry.resolve_alias(start)
3239
for start in starting_with]
3240
if test_suite_factory is None:
3241
# Reduce loading time by loading modules based on the starting_with
3243
suite = test_suite(keep_only, starting_with)
3245
suite = test_suite_factory()
3247
# But always filter as requested.
3248
suite = filter_suite_by_id_startswith(suite, starting_with)
3249
return run_suite(suite, 'testbzr', verbose=verbose, pattern=pattern,
3250
stop_on_failure=stop_on_failure,
3251
transport=transport,
3252
lsprof_timed=lsprof_timed,
3253
bench_history=bench_history,
3254
matching_tests_first=matching_tests_first,
3255
list_only=list_only,
3256
random_seed=random_seed,
3257
exclude_pattern=exclude_pattern,
3259
runner_class=runner_class,
3260
suite_decorators=suite_decorators,
251
sys.stdout = real_stdout
252
sys.stderr = real_stderr
254
_show_results(result)
256
return result.wasSuccessful()
261
def _setup_test_log():
265
log_filename = os.path.abspath('testbzr.log')
266
TestBase.TEST_LOG = open(log_filename, 'wt', buffering=1) # line buffered
268
print >>TestBase.TEST_LOG, "bzr tests run at " + time.ctime()
269
print '%-30s %s' % ('test log', log_filename)
272
def _setup_test_dir():
276
TestBase.ORIG_DIR = os.getcwdu()
277
TestBase.TEST_ROOT = os.path.abspath("testbzr.tmp")
279
print '%-30s %s' % ('running tests in', TestBase.TEST_ROOT)
281
if os.path.exists(TestBase.TEST_ROOT):
282
shutil.rmtree(TestBase.TEST_ROOT)
283
os.mkdir(TestBase.TEST_ROOT)
284
os.chdir(TestBase.TEST_ROOT)
286
# make a fake bzr directory there to prevent any tests propagating
287
# up onto the source directory's real branch
288
os.mkdir(os.path.join(TestBase.TEST_ROOT, '.bzr'))
292
def _show_results(result):
294
print '%4d tests run' % result.testsRun
295
print '%4d errors' % len(result.errors)
296
print '%4d failures' % len(result.failures)
300
def _show_test_failure(kind, case, exc_info, out):
301
from traceback import print_exception
303
print >>out, '-' * 60
306
desc = case.shortDescription()
308
print >>out, ' (%s)' % desc
310
print_exception(exc_info[0], exc_info[1], exc_info[2], None, out)
312
if isinstance(case, TestBase):
314
print >>out, 'log from this test:'
315
print >>out, case._log_buf
317
print >>out, '-' * 60
3264
default_transport = old_transport
3265
selftest_debug_flags = old_debug_flags
3268
def load_test_id_list(file_name):
3269
"""Load a test id list from a text file.
3271
The format is one test id by line. No special care is taken to impose
3272
strict rules, these test ids are used to filter the test suite so a test id
3273
that do not match an existing test will do no harm. This allows user to add
3274
comments, leave blank lines, etc.
3278
ftest = open(file_name, 'rt')
3280
if e.errno != errno.ENOENT:
3283
raise errors.NoSuchFile(file_name)
3285
for test_name in ftest.readlines():
3286
test_list.append(test_name.strip())
3291
def suite_matches_id_list(test_suite, id_list):
3292
"""Warns about tests not appearing or appearing more than once.
3294
:param test_suite: A TestSuite object.
3295
:param test_id_list: The list of test ids that should be found in
3298
:return: (absents, duplicates) absents is a list containing the test found
3299
in id_list but not in test_suite, duplicates is a list containing the
3300
test found multiple times in test_suite.
3302
When using a prefined test id list, it may occurs that some tests do not
3303
exist anymore or that some tests use the same id. This function warns the
3304
tester about potential problems in his workflow (test lists are volatile)
3305
or in the test suite itself (using the same id for several tests does not
3306
help to localize defects).
3308
# Build a dict counting id occurrences
3310
for test in iter_suite_tests(test_suite):
3312
tests[id] = tests.get(id, 0) + 1
3317
occurs = tests.get(id, 0)
3319
not_found.append(id)
3321
duplicates.append(id)
3323
return not_found, duplicates
3326
class TestIdList(object):
3327
"""Test id list to filter a test suite.
3329
Relying on the assumption that test ids are built as:
3330
<module>[.<class>.<method>][(<param>+)], <module> being in python dotted
3331
notation, this class offers methods to :
3332
- avoid building a test suite for modules not refered to in the test list,
3333
- keep only the tests listed from the module test suite.
3336
def __init__(self, test_id_list):
3337
# When a test suite needs to be filtered against us we compare test ids
3338
# for equality, so a simple dict offers a quick and simple solution.
3339
self.tests = dict().fromkeys(test_id_list, True)
3341
# While unittest.TestCase have ids like:
3342
# <module>.<class>.<method>[(<param+)],
3343
# doctest.DocTestCase can have ids like:
3346
# <module>.<function>
3347
# <module>.<class>.<method>
3349
# Since we can't predict a test class from its name only, we settle on
3350
# a simple constraint: a test id always begins with its module name.
3353
for test_id in test_id_list:
3354
parts = test_id.split('.')
3355
mod_name = parts.pop(0)
3356
modules[mod_name] = True
3358
mod_name += '.' + part
3359
modules[mod_name] = True
3360
self.modules = modules
3362
def refers_to(self, module_name):
3363
"""Is there tests for the module or one of its sub modules."""
3364
return self.modules.has_key(module_name)
3366
def includes(self, test_id):
3367
return self.tests.has_key(test_id)
3370
class TestPrefixAliasRegistry(registry.Registry):
3371
"""A registry for test prefix aliases.
3373
This helps implement shorcuts for the --starting-with selftest
3374
option. Overriding existing prefixes is not allowed but not fatal (a
3375
warning will be emitted).
3378
def register(self, key, obj, help=None, info=None,
3379
override_existing=False):
3380
"""See Registry.register.
3382
Trying to override an existing alias causes a warning to be emitted,
3383
not a fatal execption.
3386
super(TestPrefixAliasRegistry, self).register(
3387
key, obj, help=help, info=info, override_existing=False)
3389
actual = self.get(key)
3390
note('Test prefix alias %s is already used for %s, ignoring %s'
3391
% (key, actual, obj))
3393
def resolve_alias(self, id_start):
3394
"""Replace the alias by the prefix in the given string.
3396
Using an unknown prefix is an error to help catching typos.
3398
parts = id_start.split('.')
3400
parts[0] = self.get(parts[0])
3402
raise errors.BzrCommandError(
3403
'%s is not a known test prefix alias' % parts[0])
3404
return '.'.join(parts)
3407
test_prefix_alias_registry = TestPrefixAliasRegistry()
3408
"""Registry of test prefix aliases."""
3411
# This alias allows to detect typos ('bzrlin.') by making all valid test ids
3412
# appear prefixed ('bzrlib.' is "replaced" by 'bzrlib.').
3413
test_prefix_alias_registry.register('bzrlib', 'bzrlib')
3415
# Obvious higest levels prefixes, feel free to add your own via a plugin
3416
test_prefix_alias_registry.register('bd', 'bzrlib.doc')
3417
test_prefix_alias_registry.register('bu', 'bzrlib.utils')
3418
test_prefix_alias_registry.register('bt', 'bzrlib.tests')
3419
test_prefix_alias_registry.register('bb', 'bzrlib.tests.blackbox')
3420
test_prefix_alias_registry.register('bp', 'bzrlib.plugins')
3423
def test_suite(keep_only=None, starting_with=None):
3424
"""Build and return TestSuite for the whole of bzrlib.
3426
:param keep_only: A list of test ids limiting the suite returned.
3428
:param starting_with: An id limiting the suite returned to the tests
3431
This function can be replaced if you need to change the default test
3432
suite on a global basis, but it is not encouraged.
3436
'bzrlib.tests.blackbox',
3437
'bzrlib.tests.commands',
3438
'bzrlib.tests.per_branch',
3439
'bzrlib.tests.per_bzrdir',
3440
'bzrlib.tests.per_interrepository',
3441
'bzrlib.tests.per_intertree',
3442
'bzrlib.tests.per_inventory',
3443
'bzrlib.tests.per_interbranch',
3444
'bzrlib.tests.per_lock',
3445
'bzrlib.tests.per_transport',
3446
'bzrlib.tests.per_tree',
3447
'bzrlib.tests.per_pack_repository',
3448
'bzrlib.tests.per_repository',
3449
'bzrlib.tests.per_repository_chk',
3450
'bzrlib.tests.per_repository_reference',
3451
'bzrlib.tests.per_versionedfile',
3452
'bzrlib.tests.per_workingtree',
3453
'bzrlib.tests.test__annotator',
3454
'bzrlib.tests.test__chk_map',
3455
'bzrlib.tests.test__dirstate_helpers',
3456
'bzrlib.tests.test__groupcompress',
3457
'bzrlib.tests.test__known_graph',
3458
'bzrlib.tests.test__rio',
3459
'bzrlib.tests.test__walkdirs_win32',
3460
'bzrlib.tests.test_ancestry',
3461
'bzrlib.tests.test_annotate',
3462
'bzrlib.tests.test_api',
3463
'bzrlib.tests.test_atomicfile',
3464
'bzrlib.tests.test_bad_files',
3465
'bzrlib.tests.test_bencode',
3466
'bzrlib.tests.test_bisect_multi',
3467
'bzrlib.tests.test_branch',
3468
'bzrlib.tests.test_branchbuilder',
3469
'bzrlib.tests.test_btree_index',
3470
'bzrlib.tests.test_bugtracker',
3471
'bzrlib.tests.test_bundle',
3472
'bzrlib.tests.test_bzrdir',
3473
'bzrlib.tests.test__chunks_to_lines',
3474
'bzrlib.tests.test_cache_utf8',
3475
'bzrlib.tests.test_chk_map',
3476
'bzrlib.tests.test_chk_serializer',
3477
'bzrlib.tests.test_chunk_writer',
3478
'bzrlib.tests.test_clean_tree',
3479
'bzrlib.tests.test_commands',
3480
'bzrlib.tests.test_commit',
3481
'bzrlib.tests.test_commit_merge',
3482
'bzrlib.tests.test_config',
3483
'bzrlib.tests.test_conflicts',
3484
'bzrlib.tests.test_counted_lock',
3485
'bzrlib.tests.test_crash',
3486
'bzrlib.tests.test_decorators',
3487
'bzrlib.tests.test_delta',
3488
'bzrlib.tests.test_debug',
3489
'bzrlib.tests.test_deprecated_graph',
3490
'bzrlib.tests.test_diff',
3491
'bzrlib.tests.test_directory_service',
3492
'bzrlib.tests.test_dirstate',
3493
'bzrlib.tests.test_email_message',
3494
'bzrlib.tests.test_eol_filters',
3495
'bzrlib.tests.test_errors',
3496
'bzrlib.tests.test_export',
3497
'bzrlib.tests.test_extract',
3498
'bzrlib.tests.test_fetch',
3499
'bzrlib.tests.test_fifo_cache',
3500
'bzrlib.tests.test_filters',
3501
'bzrlib.tests.test_ftp_transport',
3502
'bzrlib.tests.test_foreign',
3503
'bzrlib.tests.test_generate_docs',
3504
'bzrlib.tests.test_generate_ids',
3505
'bzrlib.tests.test_globbing',
3506
'bzrlib.tests.test_gpg',
3507
'bzrlib.tests.test_graph',
3508
'bzrlib.tests.test_groupcompress',
3509
'bzrlib.tests.test_hashcache',
3510
'bzrlib.tests.test_help',
3511
'bzrlib.tests.test_hooks',
3512
'bzrlib.tests.test_http',
3513
'bzrlib.tests.test_http_response',
3514
'bzrlib.tests.test_https_ca_bundle',
3515
'bzrlib.tests.test_identitymap',
3516
'bzrlib.tests.test_ignores',
3517
'bzrlib.tests.test_index',
3518
'bzrlib.tests.test_info',
3519
'bzrlib.tests.test_inv',
3520
'bzrlib.tests.test_inventory_delta',
3521
'bzrlib.tests.test_knit',
3522
'bzrlib.tests.test_lazy_import',
3523
'bzrlib.tests.test_lazy_regex',
3524
'bzrlib.tests.test_lock',
3525
'bzrlib.tests.test_lockable_files',
3526
'bzrlib.tests.test_lockdir',
3527
'bzrlib.tests.test_log',
3528
'bzrlib.tests.test_lru_cache',
3529
'bzrlib.tests.test_lsprof',
3530
'bzrlib.tests.test_mail_client',
3531
'bzrlib.tests.test_memorytree',
3532
'bzrlib.tests.test_merge',
3533
'bzrlib.tests.test_merge3',
3534
'bzrlib.tests.test_merge_core',
3535
'bzrlib.tests.test_merge_directive',
3536
'bzrlib.tests.test_missing',
3537
'bzrlib.tests.test_msgeditor',
3538
'bzrlib.tests.test_multiparent',
3539
'bzrlib.tests.test_mutabletree',
3540
'bzrlib.tests.test_nonascii',
3541
'bzrlib.tests.test_options',
3542
'bzrlib.tests.test_osutils',
3543
'bzrlib.tests.test_osutils_encodings',
3544
'bzrlib.tests.test_pack',
3545
'bzrlib.tests.test_patch',
3546
'bzrlib.tests.test_patches',
3547
'bzrlib.tests.test_permissions',
3548
'bzrlib.tests.test_plugins',
3549
'bzrlib.tests.test_progress',
3550
'bzrlib.tests.test_read_bundle',
3551
'bzrlib.tests.test_reconcile',
3552
'bzrlib.tests.test_reconfigure',
3553
'bzrlib.tests.test_registry',
3554
'bzrlib.tests.test_remote',
3555
'bzrlib.tests.test_rename_map',
3556
'bzrlib.tests.test_repository',
3557
'bzrlib.tests.test_revert',
3558
'bzrlib.tests.test_revision',
3559
'bzrlib.tests.test_revisionspec',
3560
'bzrlib.tests.test_revisiontree',
3561
'bzrlib.tests.test_rio',
3562
'bzrlib.tests.test_rules',
3563
'bzrlib.tests.test_sampler',
3564
'bzrlib.tests.test_selftest',
3565
'bzrlib.tests.test_serializer',
3566
'bzrlib.tests.test_setup',
3567
'bzrlib.tests.test_sftp_transport',
3568
'bzrlib.tests.test_shelf',
3569
'bzrlib.tests.test_shelf_ui',
3570
'bzrlib.tests.test_smart',
3571
'bzrlib.tests.test_smart_add',
3572
'bzrlib.tests.test_smart_request',
3573
'bzrlib.tests.test_smart_transport',
3574
'bzrlib.tests.test_smtp_connection',
3575
'bzrlib.tests.test_source',
3576
'bzrlib.tests.test_ssh_transport',
3577
'bzrlib.tests.test_status',
3578
'bzrlib.tests.test_store',
3579
'bzrlib.tests.test_strace',
3580
'bzrlib.tests.test_subsume',
3581
'bzrlib.tests.test_switch',
3582
'bzrlib.tests.test_symbol_versioning',
3583
'bzrlib.tests.test_tag',
3584
'bzrlib.tests.test_testament',
3585
'bzrlib.tests.test_textfile',
3586
'bzrlib.tests.test_textmerge',
3587
'bzrlib.tests.test_timestamp',
3588
'bzrlib.tests.test_trace',
3589
'bzrlib.tests.test_transactions',
3590
'bzrlib.tests.test_transform',
3591
'bzrlib.tests.test_transport',
3592
'bzrlib.tests.test_transport_log',
3593
'bzrlib.tests.test_tree',
3594
'bzrlib.tests.test_treebuilder',
3595
'bzrlib.tests.test_tsort',
3596
'bzrlib.tests.test_tuned_gzip',
3597
'bzrlib.tests.test_ui',
3598
'bzrlib.tests.test_uncommit',
3599
'bzrlib.tests.test_upgrade',
3600
'bzrlib.tests.test_upgrade_stacked',
3601
'bzrlib.tests.test_urlutils',
3602
'bzrlib.tests.test_version',
3603
'bzrlib.tests.test_version_info',
3604
'bzrlib.tests.test_weave',
3605
'bzrlib.tests.test_whitebox',
3606
'bzrlib.tests.test_win32utils',
3607
'bzrlib.tests.test_workingtree',
3608
'bzrlib.tests.test_workingtree_4',
3609
'bzrlib.tests.test_wsgi',
3610
'bzrlib.tests.test_xml',
3613
loader = TestUtil.TestLoader()
3615
if keep_only is not None:
3616
id_filter = TestIdList(keep_only)
3618
# We take precedence over keep_only because *at loading time* using
3619
# both options means we will load less tests for the same final result.
3620
def interesting_module(name):
3621
for start in starting_with:
3623
# Either the module name starts with the specified string
3624
name.startswith(start)
3625
# or it may contain tests starting with the specified string
3626
or start.startswith(name)
3630
loader = TestUtil.FilteredByModuleTestLoader(interesting_module)
3632
elif keep_only is not None:
3633
loader = TestUtil.FilteredByModuleTestLoader(id_filter.refers_to)
3634
def interesting_module(name):
3635
return id_filter.refers_to(name)
3638
loader = TestUtil.TestLoader()
3639
def interesting_module(name):
3640
# No filtering, all modules are interesting
3643
suite = loader.suiteClass()
3645
# modules building their suite with loadTestsFromModuleNames
3646
suite.addTest(loader.loadTestsFromModuleNames(testmod_names))
3648
modules_to_doctest = [
3650
'bzrlib.branchbuilder',
3653
'bzrlib.iterablefile',
3657
'bzrlib.symbol_versioning',
3660
'bzrlib.version_info_formats.format_custom',
3663
for mod in modules_to_doctest:
3664
if not interesting_module(mod):
3665
# No tests to keep here, move along
3668
# note that this really does mean "report only" -- doctest
3669
# still runs the rest of the examples
3670
doc_suite = doctest.DocTestSuite(mod,
3671
optionflags=doctest.REPORT_ONLY_FIRST_FAILURE)
3672
except ValueError, e:
3673
print '**failed to get doctest for: %s\n%s' % (mod, e)
3675
if len(doc_suite._tests) == 0:
3676
raise errors.BzrError("no doctests found in %s" % (mod,))
3677
suite.addTest(doc_suite)
3679
default_encoding = sys.getdefaultencoding()
3680
for name, plugin in bzrlib.plugin.plugins().items():
3681
if not interesting_module(plugin.module.__name__):
3683
plugin_suite = plugin.test_suite()
3684
# We used to catch ImportError here and turn it into just a warning,
3685
# but really if you don't have --no-plugins this should be a failure.
3686
# mbp 20080213 - see http://bugs.launchpad.net/bugs/189771
3687
if plugin_suite is None:
3688
plugin_suite = plugin.load_plugin_tests(loader)
3689
if plugin_suite is not None:
3690
suite.addTest(plugin_suite)
3691
if default_encoding != sys.getdefaultencoding():
3692
bzrlib.trace.warning(
3693
'Plugin "%s" tried to reset default encoding to: %s', name,
3694
sys.getdefaultencoding())
3696
sys.setdefaultencoding(default_encoding)
3698
if keep_only is not None:
3699
# Now that the referred modules have loaded their tests, keep only the
3701
suite = filter_suite_by_id_list(suite, id_filter)
3702
# Do some sanity checks on the id_list filtering
3703
not_found, duplicates = suite_matches_id_list(suite, keep_only)
3705
# The tester has used both keep_only and starting_with, so he is
3706
# already aware that some tests are excluded from the list, there
3707
# is no need to tell him which.
3710
# Some tests mentioned in the list are not in the test suite. The
3711
# list may be out of date, report to the tester.
3712
for id in not_found:
3713
bzrlib.trace.warning('"%s" not found in the test suite', id)
3714
for id in duplicates:
3715
bzrlib.trace.warning('"%s" is used as an id by several tests', id)
3720
def multiply_scenarios(scenarios_left, scenarios_right):
3721
"""Multiply two sets of scenarios.
3723
:returns: the cartesian product of the two sets of scenarios, that is
3724
a scenario for every possible combination of a left scenario and a
3728
('%s,%s' % (left_name, right_name),
3729
dict(left_dict.items() + right_dict.items()))
3730
for left_name, left_dict in scenarios_left
3731
for right_name, right_dict in scenarios_right]
3734
def multiply_tests(tests, scenarios, result):
3735
"""Multiply tests_list by scenarios into result.
3737
This is the core workhorse for test parameterisation.
3739
Typically the load_tests() method for a per-implementation test suite will
3740
call multiply_tests and return the result.
3742
:param tests: The tests to parameterise.
3743
:param scenarios: The scenarios to apply: pairs of (scenario_name,
3744
scenario_param_dict).
3745
:param result: A TestSuite to add created tests to.
3747
This returns the passed in result TestSuite with the cross product of all
3748
the tests repeated once for each scenario. Each test is adapted by adding
3749
the scenario name at the end of its id(), and updating the test object's
3750
__dict__ with the scenario_param_dict.
3752
>>> import bzrlib.tests.test_sampler
3753
>>> r = multiply_tests(
3754
... bzrlib.tests.test_sampler.DemoTest('test_nothing'),
3755
... [('one', dict(param=1)),
3756
... ('two', dict(param=2))],
3758
>>> tests = list(iter_suite_tests(r))
3762
'bzrlib.tests.test_sampler.DemoTest.test_nothing(one)'
3768
for test in iter_suite_tests(tests):
3769
apply_scenarios(test, scenarios, result)
3773
def apply_scenarios(test, scenarios, result):
3774
"""Apply the scenarios in scenarios to test and add to result.
3776
:param test: The test to apply scenarios to.
3777
:param scenarios: An iterable of scenarios to apply to test.
3779
:seealso: apply_scenario
3781
for scenario in scenarios:
3782
result.addTest(apply_scenario(test, scenario))
3786
def apply_scenario(test, scenario):
3787
"""Copy test and apply scenario to it.
3789
:param test: A test to adapt.
3790
:param scenario: A tuple describing the scenarion.
3791
The first element of the tuple is the new test id.
3792
The second element is a dict containing attributes to set on the
3794
:return: The adapted test.
3796
new_id = "%s(%s)" % (test.id(), scenario[0])
3797
new_test = clone_test(test, new_id)
3798
for name, value in scenario[1].items():
3799
setattr(new_test, name, value)
3803
def clone_test(test, new_id):
3804
"""Clone a test giving it a new id.
3806
:param test: The test to clone.
3807
:param new_id: The id to assign to it.
3808
:return: The new test.
3810
from copy import deepcopy
3811
new_test = deepcopy(test)
3812
new_test.id = lambda: new_id
3816
def _rmtree_temp_dir(dirname):
3817
# If LANG=C we probably have created some bogus paths
3818
# which rmtree(unicode) will fail to delete
3819
# so make sure we are using rmtree(str) to delete everything
3820
# except on win32, where rmtree(str) will fail
3821
# since it doesn't have the property of byte-stream paths
3822
# (they are either ascii or mbcs)
3823
if sys.platform == 'win32':
3824
# make sure we are using the unicode win32 api
3825
dirname = unicode(dirname)
3827
dirname = dirname.encode(sys.getfilesystemencoding())
3829
osutils.rmtree(dirname)
3831
# We don't want to fail here because some useful display will be lost
3832
# otherwise. Polluting the tmp dir is bad, but not giving all the
3833
# possible info to the test runner is even worse.
3834
sys.stderr.write('Unable to remove testing dir %s\n%s'
3835
% (os.path.basename(dirname), e))
3838
class Feature(object):
3839
"""An operating system Feature."""
3842
self._available = None
3844
def available(self):
3845
"""Is the feature available?
3847
:return: True if the feature is available.
3849
if self._available is None:
3850
self._available = self._probe()
3851
return self._available
3854
"""Implement this method in concrete features.
3856
:return: True if the feature is available.
3858
raise NotImplementedError
3861
if getattr(self, 'feature_name', None):
3862
return self.feature_name()
3863
return self.__class__.__name__
3866
class _SymlinkFeature(Feature):
3869
return osutils.has_symlinks()
3871
def feature_name(self):
3874
SymlinkFeature = _SymlinkFeature()
3877
class _HardlinkFeature(Feature):
3880
return osutils.has_hardlinks()
3882
def feature_name(self):
3885
HardlinkFeature = _HardlinkFeature()
3888
class _OsFifoFeature(Feature):
3891
return getattr(os, 'mkfifo', None)
3893
def feature_name(self):
3894
return 'filesystem fifos'
3896
OsFifoFeature = _OsFifoFeature()
3899
class _UnicodeFilenameFeature(Feature):
3900
"""Does the filesystem support Unicode filenames?"""
3904
# Check for character combinations unlikely to be covered by any
3905
# single non-unicode encoding. We use the characters
3906
# - greek small letter alpha (U+03B1) and
3907
# - braille pattern dots-123456 (U+283F).
3908
os.stat(u'\u03b1\u283f')
3909
except UnicodeEncodeError:
3911
except (IOError, OSError):
3912
# The filesystem allows the Unicode filename but the file doesn't
3916
# The filesystem allows the Unicode filename and the file exists,
3920
UnicodeFilenameFeature = _UnicodeFilenameFeature()
3923
def probe_unicode_in_user_encoding():
3924
"""Try to encode several unicode strings to use in unicode-aware tests.
3925
Return first successfull match.
3927
:return: (unicode value, encoded plain string value) or (None, None)
3929
possible_vals = [u'm\xb5', u'\xe1', u'\u0410']
3930
for uni_val in possible_vals:
3932
str_val = uni_val.encode(osutils.get_user_encoding())
3933
except UnicodeEncodeError:
3934
# Try a different character
3937
return uni_val, str_val
3941
def probe_bad_non_ascii(encoding):
3942
"""Try to find [bad] character with code [128..255]
3943
that cannot be decoded to unicode in some encoding.
3944
Return None if all non-ascii characters is valid
3947
for i in xrange(128, 256):
3950
char.decode(encoding)
3951
except UnicodeDecodeError:
3956
class _HTTPSServerFeature(Feature):
3957
"""Some tests want an https Server, check if one is available.
3959
Right now, the only way this is available is under python2.6 which provides
3970
def feature_name(self):
3971
return 'HTTPSServer'
3974
HTTPSServerFeature = _HTTPSServerFeature()
3977
class _UnicodeFilename(Feature):
3978
"""Does the filesystem support Unicode filenames?"""
3983
except UnicodeEncodeError:
3985
except (IOError, OSError):
3986
# The filesystem allows the Unicode filename but the file doesn't
3990
# The filesystem allows the Unicode filename and the file exists,
3994
UnicodeFilename = _UnicodeFilename()
3997
class _UTF8Filesystem(Feature):
3998
"""Is the filesystem UTF-8?"""
4001
if osutils._fs_enc.upper() in ('UTF-8', 'UTF8'):
4005
UTF8Filesystem = _UTF8Filesystem()
4008
class _CaseInsCasePresFilenameFeature(Feature):
4009
"""Is the file-system case insensitive, but case-preserving?"""
4012
fileno, name = tempfile.mkstemp(prefix='MixedCase')
4014
# first check truly case-preserving for created files, then check
4015
# case insensitive when opening existing files.
4016
name = osutils.normpath(name)
4017
base, rel = osutils.split(name)
4018
found_rel = osutils.canonical_relpath(base, name)
4019
return (found_rel == rel
4020
and os.path.isfile(name.upper())
4021
and os.path.isfile(name.lower()))
4026
def feature_name(self):
4027
return "case-insensitive case-preserving filesystem"
4029
CaseInsCasePresFilenameFeature = _CaseInsCasePresFilenameFeature()
4032
class _CaseInsensitiveFilesystemFeature(Feature):
4033
"""Check if underlying filesystem is case-insensitive but *not* case
4036
# Note that on Windows, Cygwin, MacOS etc, the file-systems are far
4037
# more likely to be case preserving, so this case is rare.
4040
if CaseInsCasePresFilenameFeature.available():
4043
if TestCaseWithMemoryTransport.TEST_ROOT is None:
4044
root = osutils.mkdtemp(prefix='testbzr-', suffix='.tmp')
4045
TestCaseWithMemoryTransport.TEST_ROOT = root
4047
root = TestCaseWithMemoryTransport.TEST_ROOT
4048
tdir = osutils.mkdtemp(prefix='case-sensitive-probe-', suffix='',
4050
name_a = osutils.pathjoin(tdir, 'a')
4051
name_A = osutils.pathjoin(tdir, 'A')
4053
result = osutils.isdir(name_A)
4054
_rmtree_temp_dir(tdir)
4057
def feature_name(self):
4058
return 'case-insensitive filesystem'
4060
CaseInsensitiveFilesystemFeature = _CaseInsensitiveFilesystemFeature()
4063
class _SubUnitFeature(Feature):
4064
"""Check if subunit is available."""
4073
def feature_name(self):
4076
SubUnitFeature = _SubUnitFeature()
4077
# Only define SubUnitBzrRunner if subunit is available.
4079
from subunit import TestProtocolClient
4081
from subunit.test_results import AutoTimingTestResultDecorator
4083
AutoTimingTestResultDecorator = lambda x:x
4084
class SubUnitBzrRunner(TextTestRunner):
4085
def run(self, test):
4086
result = AutoTimingTestResultDecorator(
4087
TestProtocolClient(self.stream))
added
removed
bzrlib/tests/__init__.py
