1
# Copyright (C) 2005, 2006, 2007 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 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.
30
from cStringIO import StringIO
36
from pprint import pformat
40
from subprocess import Popen, PIPE
58
import bzrlib.commands
59
import bzrlib.timestamp
61
import bzrlib.inventory
62
import bzrlib.iterablefile
67
# lsprof not available
69
from bzrlib.merge import merge_inner
73
from bzrlib.revision import common_ancestor
75
from bzrlib import symbol_versioning
77
from bzrlib.transport import get_transport
78
import bzrlib.transport
79
from bzrlib.transport.local import LocalURLServer
80
from bzrlib.transport.memory import MemoryServer
81
from bzrlib.transport.readonly import ReadonlyServer
82
from bzrlib.trace import mutter, note
83
from bzrlib.tests import TestUtil
84
from bzrlib.tests.HttpServer import HttpServer
85
from bzrlib.tests.TestUtil import (
89
from bzrlib.tests.treeshape import build_tree_contents
90
from bzrlib.workingtree import WorkingTree, WorkingTreeFormat2
92
default_transport = LocalURLServer
95
MODULES_TO_DOCTEST = [
107
NUMBERED_DIRS = False # dirs kind for TestCaseInTempDir (numbered or named)
110
def packages_to_test():
111
"""Return a list of packages to test.
113
The packages are not globally imported so that import failures are
114
triggered when running selftest, not when importing the command.
117
import bzrlib.tests.blackbox
118
import bzrlib.tests.branch_implementations
119
import bzrlib.tests.bzrdir_implementations
120
import bzrlib.tests.interrepository_implementations
121
import bzrlib.tests.interversionedfile_implementations
122
import bzrlib.tests.intertree_implementations
123
import bzrlib.tests.per_lock
124
import bzrlib.tests.repository_implementations
125
import bzrlib.tests.revisionstore_implementations
126
import bzrlib.tests.tree_implementations
127
import bzrlib.tests.workingtree_implementations
130
bzrlib.tests.blackbox,
131
bzrlib.tests.branch_implementations,
132
bzrlib.tests.bzrdir_implementations,
133
bzrlib.tests.interrepository_implementations,
134
bzrlib.tests.interversionedfile_implementations,
135
bzrlib.tests.intertree_implementations,
136
bzrlib.tests.per_lock,
137
bzrlib.tests.repository_implementations,
138
bzrlib.tests.revisionstore_implementations,
139
bzrlib.tests.tree_implementations,
140
bzrlib.tests.workingtree_implementations,
144
class ExtendedTestResult(unittest._TextTestResult):
145
"""Accepts, reports and accumulates the results of running tests.
147
Compared to this unittest version this class adds support for profiling,
148
benchmarking, stopping as soon as a test fails, and skipping tests.
149
There are further-specialized subclasses for different types of display.
154
def __init__(self, stream, descriptions, verbosity,
158
"""Construct new TestResult.
160
:param bench_history: Optionally, a writable file object to accumulate
163
unittest._TextTestResult.__init__(self, stream, descriptions, verbosity)
164
if bench_history is not None:
165
from bzrlib.version import _get_bzr_source_tree
166
src_tree = _get_bzr_source_tree()
169
revision_id = src_tree.get_parent_ids()[0]
171
# XXX: if this is a brand new tree, do the same as if there
175
# XXX: If there's no branch, what should we do?
177
bench_history.write("--date %s %s\n" % (time.time(), revision_id))
178
self._bench_history = bench_history
179
self.ui = ui.ui_factory
180
self.num_tests = num_tests
182
self.failure_count = 0
183
self.known_failure_count = 0
185
self.unsupported = {}
187
self._overall_start_time = time.time()
189
def extractBenchmarkTime(self, testCase):
190
"""Add a benchmark time for the current test case."""
191
self._benchmarkTime = getattr(testCase, "_benchtime", None)
193
def _elapsedTestTimeString(self):
194
"""Return a time string for the overall time the current test has taken."""
195
return self._formatTime(time.time() - self._start_time)
197
def _testTimeString(self):
198
if self._benchmarkTime is not None:
200
self._formatTime(self._benchmarkTime),
201
self._elapsedTestTimeString())
203
return " %s" % self._elapsedTestTimeString()
205
def _formatTime(self, seconds):
206
"""Format seconds as milliseconds with leading spaces."""
207
# some benchmarks can take thousands of seconds to run, so we need 8
209
return "%8dms" % (1000 * seconds)
211
def _shortened_test_description(self, test):
213
what = re.sub(r'^bzrlib\.(tests|benchmarks)\.', '', what)
216
def startTest(self, test):
217
unittest.TestResult.startTest(self, test)
218
self.report_test_start(test)
219
test.number = self.count
220
self._recordTestStartTime()
222
def _recordTestStartTime(self):
223
"""Record that a test has started."""
224
self._start_time = time.time()
226
def _cleanupLogFile(self, test):
227
# We can only do this if we have one of our TestCases, not if
229
setKeepLogfile = getattr(test, 'setKeepLogfile', None)
230
if setKeepLogfile is not None:
233
def addError(self, test, err):
234
self.extractBenchmarkTime(test)
235
self._cleanupLogFile(test)
236
if isinstance(err[1], TestSkipped):
237
return self.addSkipped(test, err)
238
elif isinstance(err[1], UnavailableFeature):
239
return self.addNotSupported(test, err[1].args[0])
240
unittest.TestResult.addError(self, test, err)
241
self.error_count += 1
242
self.report_error(test, err)
246
def addFailure(self, test, err):
247
self._cleanupLogFile(test)
248
self.extractBenchmarkTime(test)
249
if isinstance(err[1], KnownFailure):
250
return self.addKnownFailure(test, err)
251
unittest.TestResult.addFailure(self, test, err)
252
self.failure_count += 1
253
self.report_failure(test, err)
257
def addKnownFailure(self, test, err):
258
self.known_failure_count += 1
259
self.report_known_failure(test, err)
261
def addNotSupported(self, test, feature):
262
self.unsupported.setdefault(str(feature), 0)
263
self.unsupported[str(feature)] += 1
264
self.report_unsupported(test, feature)
266
def addSuccess(self, test):
267
self.extractBenchmarkTime(test)
268
if self._bench_history is not None:
269
if self._benchmarkTime is not None:
270
self._bench_history.write("%s %s\n" % (
271
self._formatTime(self._benchmarkTime),
273
self.report_success(test)
274
unittest.TestResult.addSuccess(self, test)
276
def addSkipped(self, test, skip_excinfo):
277
self.report_skip(test, skip_excinfo)
278
# seems best to treat this as success from point-of-view of unittest
279
# -- it actually does nothing so it barely matters :)
282
except KeyboardInterrupt:
285
self.addError(test, test.__exc_info())
287
unittest.TestResult.addSuccess(self, test)
289
def printErrorList(self, flavour, errors):
290
for test, err in errors:
291
self.stream.writeln(self.separator1)
292
self.stream.write("%s: " % flavour)
294
self.stream.write('#%d ' % test.number)
295
self.stream.writeln(self.getDescription(test))
296
if getattr(test, '_get_log', None) is not None:
298
print >>self.stream, \
299
('vvvv[log from %s]' % test.id()).ljust(78,'-')
300
print >>self.stream, test._get_log()
301
print >>self.stream, \
302
('^^^^[log from %s]' % test.id()).ljust(78,'-')
303
self.stream.writeln(self.separator2)
304
self.stream.writeln("%s" % err)
309
def report_cleaning_up(self):
312
def report_success(self, test):
316
class TextTestResult(ExtendedTestResult):
317
"""Displays progress and results of tests in text form"""
319
def __init__(self, stream, descriptions, verbosity,
324
ExtendedTestResult.__init__(self, stream, descriptions, verbosity,
325
bench_history, num_tests)
327
self.pb = self.ui.nested_progress_bar()
328
self._supplied_pb = False
331
self._supplied_pb = True
332
self.pb.show_pct = False
333
self.pb.show_spinner = False
334
self.pb.show_eta = False,
335
self.pb.show_count = False
336
self.pb.show_bar = False
338
def report_starting(self):
339
self.pb.update('[test 0/%d] starting...' % (self.num_tests))
341
def _progress_prefix_text(self):
342
a = '[%d' % self.count
343
if self.num_tests is not None:
344
a +='/%d' % self.num_tests
345
a += ' in %ds' % (time.time() - self._overall_start_time)
347
a += ', %d errors' % self.error_count
348
if self.failure_count:
349
a += ', %d failed' % self.failure_count
350
if self.known_failure_count:
351
a += ', %d known failures' % self.known_failure_count
353
a += ', %d skipped' % self.skip_count
355
a += ', %d missing features' % len(self.unsupported)
359
def report_test_start(self, test):
362
self._progress_prefix_text()
364
+ self._shortened_test_description(test))
366
def _test_description(self, test):
368
return '#%d %s' % (self.count,
369
self._shortened_test_description(test))
371
return self._shortened_test_description(test)
373
def report_error(self, test, err):
374
self.pb.note('ERROR: %s\n %s\n',
375
self._test_description(test),
379
def report_failure(self, test, err):
380
self.pb.note('FAIL: %s\n %s\n',
381
self._test_description(test),
385
def report_known_failure(self, test, err):
386
self.pb.note('XFAIL: %s\n%s\n',
387
self._test_description(test), err[1])
389
def report_skip(self, test, skip_excinfo):
392
# at the moment these are mostly not things we can fix
393
# and so they just produce stipple; use the verbose reporter
396
# show test and reason for skip
397
self.pb.note('SKIP: %s\n %s\n',
398
self._shortened_test_description(test),
401
# since the class name was left behind in the still-visible
403
self.pb.note('SKIP: %s', skip_excinfo[1])
405
def report_unsupported(self, test, feature):
406
"""test cannot be run because feature is missing."""
408
def report_cleaning_up(self):
409
self.pb.update('cleaning up...')
412
if not self._supplied_pb:
416
class VerboseTestResult(ExtendedTestResult):
417
"""Produce long output, with one line per test run plus times"""
419
def _ellipsize_to_right(self, a_string, final_width):
420
"""Truncate and pad a string, keeping the right hand side"""
421
if len(a_string) > final_width:
422
result = '...' + a_string[3-final_width:]
425
return result.ljust(final_width)
427
def report_starting(self):
428
self.stream.write('running %d tests...\n' % self.num_tests)
430
def report_test_start(self, test):
432
name = self._shortened_test_description(test)
433
# width needs space for 6 char status, plus 1 for slash, plus 2 10-char
434
# numbers, plus a trailing blank
435
# when NUMBERED_DIRS: plus 5 chars on test number, plus 1 char on space
437
self.stream.write('%5d ' % self.count)
438
self.stream.write(self._ellipsize_to_right(name,
439
osutils.terminal_width()-36))
441
self.stream.write(self._ellipsize_to_right(name,
442
osutils.terminal_width()-30))
445
def _error_summary(self, err):
449
return '%s%s' % (indent, err[1])
451
def report_error(self, test, err):
452
self.stream.writeln('ERROR %s\n%s'
453
% (self._testTimeString(),
454
self._error_summary(err)))
456
def report_failure(self, test, err):
457
self.stream.writeln(' FAIL %s\n%s'
458
% (self._testTimeString(),
459
self._error_summary(err)))
461
def report_known_failure(self, test, err):
462
self.stream.writeln('XFAIL %s\n%s'
463
% (self._testTimeString(),
464
self._error_summary(err)))
466
def report_success(self, test):
467
self.stream.writeln(' OK %s' % self._testTimeString())
468
for bench_called, stats in getattr(test, '_benchcalls', []):
469
self.stream.writeln('LSProf output for %s(%s, %s)' % bench_called)
470
stats.pprint(file=self.stream)
471
# flush the stream so that we get smooth output. This verbose mode is
472
# used to show the output in PQM.
475
def report_skip(self, test, skip_excinfo):
477
self.stream.writeln(' SKIP %s\n%s'
478
% (self._testTimeString(),
479
self._error_summary(skip_excinfo)))
481
def report_unsupported(self, test, feature):
482
"""test cannot be run because feature is missing."""
483
self.stream.writeln("NODEP %s\n The feature '%s' is not available."
484
%(self._testTimeString(), feature))
488
class TextTestRunner(object):
489
stop_on_failure = False
497
self.stream = unittest._WritelnDecorator(stream)
498
self.descriptions = descriptions
499
self.verbosity = verbosity
500
self.keep_output = keep_output
501
self._bench_history = bench_history
504
"Run the given test case or test suite."
505
startTime = time.time()
506
if self.verbosity == 1:
507
result_class = TextTestResult
508
elif self.verbosity >= 2:
509
result_class = VerboseTestResult
510
result = result_class(self.stream,
513
bench_history=self._bench_history,
514
num_tests=test.countTestCases(),
516
result.stop_early = self.stop_on_failure
517
result.report_starting()
519
stopTime = time.time()
520
timeTaken = stopTime - startTime
522
self.stream.writeln(result.separator2)
523
run = result.testsRun
524
self.stream.writeln("Ran %d test%s in %.3fs" %
525
(run, run != 1 and "s" or "", timeTaken))
526
self.stream.writeln()
527
if not result.wasSuccessful():
528
self.stream.write("FAILED (")
529
failed, errored = map(len, (result.failures, result.errors))
531
self.stream.write("failures=%d" % failed)
533
if failed: self.stream.write(", ")
534
self.stream.write("errors=%d" % errored)
535
if result.known_failure_count:
536
if failed or errored: self.stream.write(", ")
537
self.stream.write("known_failure_count=%d" %
538
result.known_failure_count)
539
self.stream.writeln(")")
541
if result.known_failure_count:
542
self.stream.writeln("OK (known_failures=%d)" %
543
result.known_failure_count)
545
self.stream.writeln("OK")
546
if result.skip_count > 0:
547
skipped = result.skip_count
548
self.stream.writeln('%d test%s skipped' %
549
(skipped, skipped != 1 and "s" or ""))
550
if result.unsupported:
551
for feature, count in sorted(result.unsupported.items()):
552
self.stream.writeln("Missing feature '%s' skipped %d tests." %
554
result.report_cleaning_up()
555
# This is still a little bogus,
556
# but only a little. Folk not using our testrunner will
557
# have to delete their temp directories themselves.
558
test_root = TestCaseWithMemoryTransport.TEST_ROOT
559
if result.wasSuccessful() or not self.keep_output:
560
if test_root is not None:
561
# If LANG=C we probably have created some bogus paths
562
# which rmtree(unicode) will fail to delete
563
# so make sure we are using rmtree(str) to delete everything
564
# except on win32, where rmtree(str) will fail
565
# since it doesn't have the property of byte-stream paths
566
# (they are either ascii or mbcs)
567
if sys.platform == 'win32':
568
# make sure we are using the unicode win32 api
569
test_root = unicode(test_root)
571
test_root = test_root.encode(
572
sys.getfilesystemencoding())
574
osutils.rmtree(test_root)
576
if sys.platform == 'win32' and e.errno == errno.EACCES:
577
print >>sys.stderr, ('Permission denied: '
578
'unable to remove testing dir '
579
'%s' % os.path.basename(test_root))
583
note("Failed tests working directories are in '%s'\n", test_root)
584
TestCaseWithMemoryTransport.TEST_ROOT = None
589
def iter_suite_tests(suite):
590
"""Return all tests in a suite, recursing through nested suites"""
591
for item in suite._tests:
592
if isinstance(item, unittest.TestCase):
594
elif isinstance(item, unittest.TestSuite):
595
for r in iter_suite_tests(item):
598
raise Exception('unknown object %r inside test suite %r'
602
class TestSkipped(Exception):
603
"""Indicates that a test was intentionally skipped, rather than failing."""
606
class KnownFailure(AssertionError):
607
"""Indicates that a test failed in a precisely expected manner.
609
Such failures dont block the whole test suite from passing because they are
610
indicators of partially completed code or of future work. We have an
611
explicit error for them so that we can ensure that they are always visible:
612
KnownFailures are always shown in the output of bzr selftest.
616
class UnavailableFeature(Exception):
617
"""A feature required for this test was not available.
619
The feature should be used to construct the exception.
623
class CommandFailed(Exception):
627
class StringIOWrapper(object):
628
"""A wrapper around cStringIO which just adds an encoding attribute.
630
Internally we can check sys.stdout to see what the output encoding
631
should be. However, cStringIO has no encoding attribute that we can
632
set. So we wrap it instead.
637
def __init__(self, s=None):
639
self.__dict__['_cstring'] = StringIO(s)
641
self.__dict__['_cstring'] = StringIO()
643
def __getattr__(self, name, getattr=getattr):
644
return getattr(self.__dict__['_cstring'], name)
646
def __setattr__(self, name, val):
647
if name == 'encoding':
648
self.__dict__['encoding'] = val
650
return setattr(self._cstring, name, val)
653
class TestUIFactory(ui.CLIUIFactory):
654
"""A UI Factory for testing.
656
Hide the progress bar but emit note()s.
658
Allows get_password to be tested without real tty attached.
665
super(TestUIFactory, self).__init__()
666
if stdin is not None:
667
# We use a StringIOWrapper to be able to test various
668
# encodings, but the user is still responsible to
669
# encode the string and to set the encoding attribute
670
# of StringIOWrapper.
671
self.stdin = StringIOWrapper(stdin)
673
self.stdout = sys.stdout
677
self.stderr = sys.stderr
682
"""See progress.ProgressBar.clear()."""
684
def clear_term(self):
685
"""See progress.ProgressBar.clear_term()."""
687
def clear_term(self):
688
"""See progress.ProgressBar.clear_term()."""
691
"""See progress.ProgressBar.finished()."""
693
def note(self, fmt_string, *args, **kwargs):
694
"""See progress.ProgressBar.note()."""
695
self.stdout.write((fmt_string + "\n") % args)
697
def progress_bar(self):
700
def nested_progress_bar(self):
703
def update(self, message, count=None, total=None):
704
"""See progress.ProgressBar.update()."""
706
def get_non_echoed_password(self, prompt):
707
"""Get password from stdin without trying to handle the echo mode"""
709
self.stdout.write(prompt)
710
password = self.stdin.readline()
713
if password[-1] == '\n':
714
password = password[:-1]
718
class TestCase(unittest.TestCase):
719
"""Base class for bzr unit tests.
721
Tests that need access to disk resources should subclass
722
TestCaseInTempDir not TestCase.
724
Error and debug log messages are redirected from their usual
725
location into a temporary file, the contents of which can be
726
retrieved by _get_log(). We use a real OS file, not an in-memory object,
727
so that it can also capture file IO. When the test completes this file
728
is read into memory and removed from disk.
730
There are also convenience functions to invoke bzr's command-line
731
routine, and to build and check bzr trees.
733
In addition to the usual method of overriding tearDown(), this class also
734
allows subclasses to register functions into the _cleanups list, which is
735
run in order as the object is torn down. It's less likely this will be
736
accidentally overlooked.
739
_log_file_name = None
741
_keep_log_file = False
742
# record lsprof data when performing benchmark calls.
743
_gather_lsprof_in_benchmarks = False
745
def __init__(self, methodName='testMethod'):
746
super(TestCase, self).__init__(methodName)
750
unittest.TestCase.setUp(self)
751
self._cleanEnvironment()
752
bzrlib.trace.disable_default_logging()
755
self._benchcalls = []
756
self._benchtime = None
757
# prevent hooks affecting tests
758
self._preserved_hooks = {
759
bzrlib.branch.Branch:bzrlib.branch.Branch.hooks,
760
bzrlib.smart.server.SmartTCPServer:bzrlib.smart.server.SmartTCPServer.hooks,
762
self.addCleanup(self._restoreHooks)
763
# this list of hooks must be kept in sync with the defaults
765
bzrlib.branch.Branch.hooks = bzrlib.branch.BranchHooks()
767
def _silenceUI(self):
768
"""Turn off UI for duration of test"""
769
# by default the UI is off; tests can turn it on if they want it.
770
saved = ui.ui_factory
772
ui.ui_factory = saved
773
ui.ui_factory = ui.SilentUIFactory()
774
self.addCleanup(_restore)
776
def _ndiff_strings(self, a, b):
777
"""Return ndiff between two strings containing lines.
779
A trailing newline is added if missing to make the strings
781
if b and b[-1] != '\n':
783
if a and a[-1] != '\n':
785
difflines = difflib.ndiff(a.splitlines(True),
787
linejunk=lambda x: False,
788
charjunk=lambda x: False)
789
return ''.join(difflines)
791
def assertEqual(self, a, b, message=''):
795
except UnicodeError, e:
796
# If we can't compare without getting a UnicodeError, then
797
# obviously they are different
798
mutter('UnicodeError: %s', e)
801
raise AssertionError("%snot equal:\na = %s\nb = %s\n"
803
pformat(a, indent=4), pformat(b, indent=4)))
805
assertEquals = assertEqual
807
def assertEqualDiff(self, a, b, message=None):
808
"""Assert two texts are equal, if not raise an exception.
810
This is intended for use with multi-line strings where it can
811
be hard to find the differences by eye.
813
# TODO: perhaps override assertEquals to call this for strings?
817
message = "texts not equal:\n"
818
raise AssertionError(message +
819
self._ndiff_strings(a, b))
821
def assertEqualMode(self, mode, mode_test):
822
self.assertEqual(mode, mode_test,
823
'mode mismatch %o != %o' % (mode, mode_test))
825
def assertStartsWith(self, s, prefix):
826
if not s.startswith(prefix):
827
raise AssertionError('string %r does not start with %r' % (s, prefix))
829
def assertEndsWith(self, s, suffix):
830
"""Asserts that s ends with suffix."""
831
if not s.endswith(suffix):
832
raise AssertionError('string %r does not end with %r' % (s, suffix))
834
def assertContainsRe(self, haystack, needle_re):
835
"""Assert that a contains something matching a regular expression."""
836
if not re.search(needle_re, haystack):
837
raise AssertionError('pattern "%r" not found in "%r"'
838
% (needle_re, haystack))
840
def assertNotContainsRe(self, haystack, needle_re):
841
"""Assert that a does not match a regular expression"""
842
if re.search(needle_re, haystack):
843
raise AssertionError('pattern "%s" found in "%s"'
844
% (needle_re, haystack))
846
def assertSubset(self, sublist, superlist):
847
"""Assert that every entry in sublist is present in superlist."""
849
for entry in sublist:
850
if entry not in superlist:
851
missing.append(entry)
853
raise AssertionError("value(s) %r not present in container %r" %
854
(missing, superlist))
856
def assertListRaises(self, excClass, func, *args, **kwargs):
857
"""Fail unless excClass is raised when the iterator from func is used.
859
Many functions can return generators this makes sure
860
to wrap them in a list() call to make sure the whole generator
861
is run, and that the proper exception is raised.
864
list(func(*args, **kwargs))
868
if getattr(excClass,'__name__', None) is not None:
869
excName = excClass.__name__
871
excName = str(excClass)
872
raise self.failureException, "%s not raised" % excName
874
def assertRaises(self, excClass, func, *args, **kwargs):
875
"""Assert that a callable raises a particular exception.
877
:param excClass: As for the except statement, this may be either an
878
exception class, or a tuple of classes.
880
Returns the exception so that you can examine it.
883
func(*args, **kwargs)
887
if getattr(excClass,'__name__', None) is not None:
888
excName = excClass.__name__
891
excName = str(excClass)
892
raise self.failureException, "%s not raised" % excName
894
def assertIs(self, left, right, message=None):
895
if not (left is right):
896
if message is not None:
897
raise AssertionError(message)
899
raise AssertionError("%r is not %r." % (left, right))
901
def assertIsNot(self, left, right, message=None):
903
if message is not None:
904
raise AssertionError(message)
906
raise AssertionError("%r is %r." % (left, right))
908
def assertTransportMode(self, transport, path, mode):
909
"""Fail if a path does not have mode mode.
911
If modes are not supported on this transport, the assertion is ignored.
913
if not transport._can_roundtrip_unix_modebits():
915
path_stat = transport.stat(path)
916
actual_mode = stat.S_IMODE(path_stat.st_mode)
917
self.assertEqual(mode, actual_mode,
918
'mode of %r incorrect (%o != %o)' % (path, mode, actual_mode))
920
def assertIsInstance(self, obj, kls):
921
"""Fail if obj is not an instance of kls"""
922
if not isinstance(obj, kls):
923
self.fail("%r is an instance of %s rather than %s" % (
924
obj, obj.__class__, kls))
926
def expectFailure(self, reason, assertion, *args, **kwargs):
927
"""Invoke a test, expecting it to fail for the given reason.
929
This is for assertions that ought to succeed, but currently fail.
930
(The failure is *expected* but not *wanted*.) Please be very precise
931
about the failure you're expecting. If a new bug is introduced,
932
AssertionError should be raised, not KnownFailure.
934
Frequently, expectFailure should be followed by an opposite assertion.
937
Intended to be used with a callable that raises AssertionError as the
938
'assertion' parameter. args and kwargs are passed to the 'assertion'.
940
Raises KnownFailure if the test fails. Raises AssertionError if the
945
self.expectFailure('Math is broken', self.assertNotEqual, 54,
947
self.assertEqual(42, dynamic_val)
949
This means that a dynamic_val of 54 will cause the test to raise
950
a KnownFailure. Once math is fixed and the expectFailure is removed,
951
only a dynamic_val of 42 will allow the test to pass. Anything other
952
than 54 or 42 will cause an AssertionError.
955
assertion(*args, **kwargs)
956
except AssertionError:
957
raise KnownFailure(reason)
959
self.fail('Unexpected success. Should have failed: %s' % reason)
961
def _capture_warnings(self, a_callable, *args, **kwargs):
962
"""A helper for callDeprecated and applyDeprecated.
964
:param a_callable: A callable to call.
965
:param args: The positional arguments for the callable
966
:param kwargs: The keyword arguments for the callable
967
:return: A tuple (warnings, result). result is the result of calling
968
a_callable(*args, **kwargs).
971
def capture_warnings(msg, cls=None, stacklevel=None):
972
# we've hooked into a deprecation specific callpath,
973
# only deprecations should getting sent via it.
974
self.assertEqual(cls, DeprecationWarning)
975
local_warnings.append(msg)
976
original_warning_method = symbol_versioning.warn
977
symbol_versioning.set_warning_method(capture_warnings)
979
result = a_callable(*args, **kwargs)
981
symbol_versioning.set_warning_method(original_warning_method)
982
return (local_warnings, result)
984
def applyDeprecated(self, deprecation_format, a_callable, *args, **kwargs):
985
"""Call a deprecated callable without warning the user.
987
:param deprecation_format: The deprecation format that the callable
988
should have been deprecated with. This is the same type as the
989
parameter to deprecated_method/deprecated_function. If the
990
callable is not deprecated with this format, an assertion error
992
:param a_callable: A callable to call. This may be a bound method or
993
a regular function. It will be called with *args and **kwargs.
994
:param args: The positional arguments for the callable
995
:param kwargs: The keyword arguments for the callable
996
:return: The result of a_callable(*args, **kwargs)
998
call_warnings, result = self._capture_warnings(a_callable,
1000
expected_first_warning = symbol_versioning.deprecation_string(
1001
a_callable, deprecation_format)
1002
if len(call_warnings) == 0:
1003
self.fail("No deprecation warning generated by call to %s" %
1005
self.assertEqual(expected_first_warning, call_warnings[0])
1008
def callDeprecated(self, expected, callable, *args, **kwargs):
1009
"""Assert that a callable is deprecated in a particular way.
1011
This is a very precise test for unusual requirements. The
1012
applyDeprecated helper function is probably more suited for most tests
1013
as it allows you to simply specify the deprecation format being used
1014
and will ensure that that is issued for the function being called.
1016
:param expected: a list of the deprecation warnings expected, in order
1017
:param callable: The callable to call
1018
:param args: The positional arguments for the callable
1019
:param kwargs: The keyword arguments for the callable
1021
call_warnings, result = self._capture_warnings(callable,
1023
self.assertEqual(expected, call_warnings)
1026
def _startLogFile(self):
1027
"""Send bzr and test log messages to a temporary file.
1029
The file is removed as the test is torn down.
1031
fileno, name = tempfile.mkstemp(suffix='.log', prefix='testbzr')
1032
self._log_file = os.fdopen(fileno, 'w+')
1033
self._log_nonce = bzrlib.trace.enable_test_log(self._log_file)
1034
self._log_file_name = name
1035
self.addCleanup(self._finishLogFile)
1037
def _finishLogFile(self):
1038
"""Finished with the log file.
1040
Close the file and delete it, unless setKeepLogfile was called.
1042
if self._log_file is None:
1044
bzrlib.trace.disable_test_log(self._log_nonce)
1045
self._log_file.close()
1046
self._log_file = None
1047
if not self._keep_log_file:
1048
os.remove(self._log_file_name)
1049
self._log_file_name = None
1051
def setKeepLogfile(self):
1052
"""Make the logfile not be deleted when _finishLogFile is called."""
1053
self._keep_log_file = True
1055
def addCleanup(self, callable):
1056
"""Arrange to run a callable when this case is torn down.
1058
Callables are run in the reverse of the order they are registered,
1059
ie last-in first-out.
1061
if callable in self._cleanups:
1062
raise ValueError("cleanup function %r already registered on %s"
1064
self._cleanups.append(callable)
1066
def _cleanEnvironment(self):
1068
'BZR_HOME': None, # Don't inherit BZR_HOME to all the tests.
1069
'HOME': os.getcwd(),
1070
'APPDATA': None, # bzr now use Win32 API and don't rely on APPDATA
1072
'BZREMAIL': None, # may still be present in the environment
1074
'BZR_PROGRESS_BAR': None,
1078
'https_proxy': None,
1079
'HTTPS_PROXY': None,
1084
# Nobody cares about these ones AFAIK. So far at
1085
# least. If you do (care), please update this comment
1091
self.addCleanup(self._restoreEnvironment)
1092
for name, value in new_env.iteritems():
1093
self._captureVar(name, value)
1095
def _captureVar(self, name, newvalue):
1096
"""Set an environment variable, and reset it when finished."""
1097
self.__old_env[name] = osutils.set_or_unset_env(name, newvalue)
1099
def _restoreEnvironment(self):
1100
for name, value in self.__old_env.iteritems():
1101
osutils.set_or_unset_env(name, value)
1103
def _restoreHooks(self):
1104
for klass, hooks in self._preserved_hooks.items():
1105
setattr(klass, 'hooks', hooks)
1107
def knownFailure(self, reason):
1108
"""This test has failed for some known reason."""
1109
raise KnownFailure(reason)
1111
def run(self, result=None):
1112
if result is None: result = self.defaultTestResult()
1113
for feature in getattr(self, '_test_needs_features', []):
1114
if not feature.available():
1115
result.startTest(self)
1116
if getattr(result, 'addNotSupported', None):
1117
result.addNotSupported(self, feature)
1119
result.addSuccess(self)
1120
result.stopTest(self)
1122
return unittest.TestCase.run(self, result)
1126
unittest.TestCase.tearDown(self)
1128
def time(self, callable, *args, **kwargs):
1129
"""Run callable and accrue the time it takes to the benchmark time.
1131
If lsprofiling is enabled (i.e. by --lsprof-time to bzr selftest) then
1132
this will cause lsprofile statistics to be gathered and stored in
1135
if self._benchtime is None:
1139
if not self._gather_lsprof_in_benchmarks:
1140
return callable(*args, **kwargs)
1142
# record this benchmark
1143
ret, stats = bzrlib.lsprof.profile(callable, *args, **kwargs)
1145
self._benchcalls.append(((callable, args, kwargs), stats))
1148
self._benchtime += time.time() - start
1150
def _runCleanups(self):
1151
"""Run registered cleanup functions.
1153
This should only be called from TestCase.tearDown.
1155
# TODO: Perhaps this should keep running cleanups even if
1156
# one of them fails?
1158
# Actually pop the cleanups from the list so tearDown running
1159
# twice is safe (this happens for skipped tests).
1160
while self._cleanups:
1161
self._cleanups.pop()()
1163
def log(self, *args):
1166
def _get_log(self, keep_log_file=False):
1167
"""Return as a string the log for this test. If the file is still
1168
on disk and keep_log_file=False, delete the log file and store the
1169
content in self._log_contents."""
1170
# flush the log file, to get all content
1172
bzrlib.trace._trace_file.flush()
1173
if self._log_contents:
1174
return self._log_contents
1175
if self._log_file_name is not None:
1176
logfile = open(self._log_file_name)
1178
log_contents = logfile.read()
1181
if not keep_log_file:
1182
self._log_contents = log_contents
1184
os.remove(self._log_file_name)
1186
if sys.platform == 'win32' and e.errno == errno.EACCES:
1187
print >>sys.stderr, ('Unable to delete log file '
1188
' %r' % self._log_file_name)
1193
return "DELETED log file to reduce memory footprint"
1195
def capture(self, cmd, retcode=0):
1196
"""Shortcut that splits cmd into words, runs, and returns stdout"""
1197
return self.run_bzr_captured(cmd.split(), retcode=retcode)[0]
1199
def requireFeature(self, feature):
1200
"""This test requires a specific feature is available.
1202
:raises UnavailableFeature: When feature is not available.
1204
if not feature.available():
1205
raise UnavailableFeature(feature)
1207
def run_bzr_captured(self, argv, retcode=0, encoding=None, stdin=None,
1209
"""Invoke bzr and return (stdout, stderr).
1211
Useful for code that wants to check the contents of the
1212
output, the way error messages are presented, etc.
1214
This should be the main method for tests that want to exercise the
1215
overall behavior of the bzr application (rather than a unit test
1216
or a functional test of the library.)
1218
Much of the old code runs bzr by forking a new copy of Python, but
1219
that is slower, harder to debug, and generally not necessary.
1221
This runs bzr through the interface that catches and reports
1222
errors, and with logging set to something approximating the
1223
default, so that error reporting can be checked.
1225
:param argv: arguments to invoke bzr
1226
:param retcode: expected return code, or None for don't-care.
1227
:param encoding: encoding for sys.stdout and sys.stderr
1228
:param stdin: A string to be used as stdin for the command.
1229
:param working_dir: Change to this directory before running
1231
if encoding is None:
1232
encoding = bzrlib.user_encoding
1233
stdout = StringIOWrapper()
1234
stderr = StringIOWrapper()
1235
stdout.encoding = encoding
1236
stderr.encoding = encoding
1238
self.log('run bzr: %r', argv)
1239
# FIXME: don't call into logging here
1240
handler = logging.StreamHandler(stderr)
1241
handler.setLevel(logging.INFO)
1242
logger = logging.getLogger('')
1243
logger.addHandler(handler)
1244
old_ui_factory = ui.ui_factory
1245
ui.ui_factory = TestUIFactory(stdin=stdin, stdout=stdout, stderr=stderr)
1248
if working_dir is not None:
1249
cwd = osutils.getcwd()
1250
os.chdir(working_dir)
1253
saved_debug_flags = frozenset(debug.debug_flags)
1254
debug.debug_flags.clear()
1256
result = self.apply_redirected(ui.ui_factory.stdin,
1258
bzrlib.commands.run_bzr_catch_errors,
1261
debug.debug_flags.update(saved_debug_flags)
1263
logger.removeHandler(handler)
1264
ui.ui_factory = old_ui_factory
1268
out = stdout.getvalue()
1269
err = stderr.getvalue()
1271
self.log('output:\n%r', out)
1273
self.log('errors:\n%r', err)
1274
if retcode is not None:
1275
self.assertEquals(retcode, result)
1278
def run_bzr(self, *args, **kwargs):
1279
"""Invoke bzr, as if it were run from the command line.
1281
This should be the main method for tests that want to exercise the
1282
overall behavior of the bzr application (rather than a unit test
1283
or a functional test of the library.)
1285
This sends the stdout/stderr results into the test's log,
1286
where it may be useful for debugging. See also run_captured.
1288
:param stdin: A string to be used as stdin for the command.
1289
:param retcode: The status code the command should return
1290
:param working_dir: The directory to run the command in
1292
retcode = kwargs.pop('retcode', 0)
1293
encoding = kwargs.pop('encoding', None)
1294
stdin = kwargs.pop('stdin', None)
1295
working_dir = kwargs.pop('working_dir', None)
1296
return self.run_bzr_captured(args, retcode=retcode, encoding=encoding,
1297
stdin=stdin, working_dir=working_dir)
1299
def run_bzr_decode(self, *args, **kwargs):
1300
if 'encoding' in kwargs:
1301
encoding = kwargs['encoding']
1303
encoding = bzrlib.user_encoding
1304
return self.run_bzr(*args, **kwargs)[0].decode(encoding)
1306
def run_bzr_error(self, error_regexes, *args, **kwargs):
1307
"""Run bzr, and check that stderr contains the supplied regexes
1309
:param error_regexes: Sequence of regular expressions which
1310
must each be found in the error output. The relative ordering
1312
:param args: command-line arguments for bzr
1313
:param kwargs: Keyword arguments which are interpreted by run_bzr
1314
This function changes the default value of retcode to be 3,
1315
since in most cases this is run when you expect bzr to fail.
1316
:return: (out, err) The actual output of running the command (in case you
1317
want to do more inspection)
1320
# Make sure that commit is failing because there is nothing to do
1321
self.run_bzr_error(['no changes to commit'],
1322
'commit', '-m', 'my commit comment')
1323
# Make sure --strict is handling an unknown file, rather than
1324
# giving us the 'nothing to do' error
1325
self.build_tree(['unknown'])
1326
self.run_bzr_error(['Commit refused because there are unknown files'],
1327
'commit', '--strict', '-m', 'my commit comment')
1329
kwargs.setdefault('retcode', 3)
1330
out, err = self.run_bzr(*args, **kwargs)
1331
for regex in error_regexes:
1332
self.assertContainsRe(err, regex)
1335
def run_bzr_subprocess(self, *args, **kwargs):
1336
"""Run bzr in a subprocess for testing.
1338
This starts a new Python interpreter and runs bzr in there.
1339
This should only be used for tests that have a justifiable need for
1340
this isolation: e.g. they are testing startup time, or signal
1341
handling, or early startup code, etc. Subprocess code can't be
1342
profiled or debugged so easily.
1344
:param retcode: The status code that is expected. Defaults to 0. If
1345
None is supplied, the status code is not checked.
1346
:param env_changes: A dictionary which lists changes to environment
1347
variables. A value of None will unset the env variable.
1348
The values must be strings. The change will only occur in the
1349
child, so you don't need to fix the environment after running.
1350
:param universal_newlines: Convert CRLF => LF
1351
:param allow_plugins: By default the subprocess is run with
1352
--no-plugins to ensure test reproducibility. Also, it is possible
1353
for system-wide plugins to create unexpected output on stderr,
1354
which can cause unnecessary test failures.
1356
env_changes = kwargs.get('env_changes', {})
1357
working_dir = kwargs.get('working_dir', None)
1358
allow_plugins = kwargs.get('allow_plugins', False)
1359
process = self.start_bzr_subprocess(args, env_changes=env_changes,
1360
working_dir=working_dir,
1361
allow_plugins=allow_plugins)
1362
# We distinguish between retcode=None and retcode not passed.
1363
supplied_retcode = kwargs.get('retcode', 0)
1364
return self.finish_bzr_subprocess(process, retcode=supplied_retcode,
1365
universal_newlines=kwargs.get('universal_newlines', False),
1368
def start_bzr_subprocess(self, process_args, env_changes=None,
1369
skip_if_plan_to_signal=False,
1371
allow_plugins=False):
1372
"""Start bzr in a subprocess for testing.
1374
This starts a new Python interpreter and runs bzr in there.
1375
This should only be used for tests that have a justifiable need for
1376
this isolation: e.g. they are testing startup time, or signal
1377
handling, or early startup code, etc. Subprocess code can't be
1378
profiled or debugged so easily.
1380
:param process_args: a list of arguments to pass to the bzr executable,
1381
for example `['--version']`.
1382
:param env_changes: A dictionary which lists changes to environment
1383
variables. A value of None will unset the env variable.
1384
The values must be strings. The change will only occur in the
1385
child, so you don't need to fix the environment after running.
1386
:param skip_if_plan_to_signal: raise TestSkipped when true and os.kill
1388
:param allow_plugins: If False (default) pass --no-plugins to bzr.
1390
:returns: Popen object for the started process.
1392
if skip_if_plan_to_signal:
1393
if not getattr(os, 'kill', None):
1394
raise TestSkipped("os.kill not available.")
1396
if env_changes is None:
1400
def cleanup_environment():
1401
for env_var, value in env_changes.iteritems():
1402
old_env[env_var] = osutils.set_or_unset_env(env_var, value)
1404
def restore_environment():
1405
for env_var, value in old_env.iteritems():
1406
osutils.set_or_unset_env(env_var, value)
1408
bzr_path = self.get_bzr_path()
1411
if working_dir is not None:
1412
cwd = osutils.getcwd()
1413
os.chdir(working_dir)
1416
# win32 subprocess doesn't support preexec_fn
1417
# so we will avoid using it on all platforms, just to
1418
# make sure the code path is used, and we don't break on win32
1419
cleanup_environment()
1420
command = [sys.executable, bzr_path]
1421
if not allow_plugins:
1422
command.append('--no-plugins')
1423
command.extend(process_args)
1424
process = self._popen(command, stdin=PIPE, stdout=PIPE, stderr=PIPE)
1426
restore_environment()
1432
def _popen(self, *args, **kwargs):
1433
"""Place a call to Popen.
1435
Allows tests to override this method to intercept the calls made to
1436
Popen for introspection.
1438
return Popen(*args, **kwargs)
1440
def get_bzr_path(self):
1441
"""Return the path of the 'bzr' executable for this test suite."""
1442
bzr_path = os.path.dirname(os.path.dirname(bzrlib.__file__))+'/bzr'
1443
if not os.path.isfile(bzr_path):
1444
# We are probably installed. Assume sys.argv is the right file
1445
bzr_path = sys.argv[0]
1448
def finish_bzr_subprocess(self, process, retcode=0, send_signal=None,
1449
universal_newlines=False, process_args=None):
1450
"""Finish the execution of process.
1452
:param process: the Popen object returned from start_bzr_subprocess.
1453
:param retcode: The status code that is expected. Defaults to 0. If
1454
None is supplied, the status code is not checked.
1455
:param send_signal: an optional signal to send to the process.
1456
:param universal_newlines: Convert CRLF => LF
1457
:returns: (stdout, stderr)
1459
if send_signal is not None:
1460
os.kill(process.pid, send_signal)
1461
out, err = process.communicate()
1463
if universal_newlines:
1464
out = out.replace('\r\n', '\n')
1465
err = err.replace('\r\n', '\n')
1467
if retcode is not None and retcode != process.returncode:
1468
if process_args is None:
1469
process_args = "(unknown args)"
1470
mutter('Output of bzr %s:\n%s', process_args, out)
1471
mutter('Error for bzr %s:\n%s', process_args, err)
1472
self.fail('Command bzr %s failed with retcode %s != %s'
1473
% (process_args, retcode, process.returncode))
1476
def check_inventory_shape(self, inv, shape):
1477
"""Compare an inventory to a list of expected names.
1479
Fail if they are not precisely equal.
1482
shape = list(shape) # copy
1483
for path, ie in inv.entries():
1484
name = path.replace('\\', '/')
1485
if ie.kind == 'dir':
1492
self.fail("expected paths not found in inventory: %r" % shape)
1494
self.fail("unexpected paths found in inventory: %r" % extras)
1496
def apply_redirected(self, stdin=None, stdout=None, stderr=None,
1497
a_callable=None, *args, **kwargs):
1498
"""Call callable with redirected std io pipes.
1500
Returns the return code."""
1501
if not callable(a_callable):
1502
raise ValueError("a_callable must be callable.")
1504
stdin = StringIO("")
1506
if getattr(self, "_log_file", None) is not None:
1507
stdout = self._log_file
1511
if getattr(self, "_log_file", None is not None):
1512
stderr = self._log_file
1515
real_stdin = sys.stdin
1516
real_stdout = sys.stdout
1517
real_stderr = sys.stderr
1522
return a_callable(*args, **kwargs)
1524
sys.stdout = real_stdout
1525
sys.stderr = real_stderr
1526
sys.stdin = real_stdin
1528
@symbol_versioning.deprecated_method(symbol_versioning.zero_eleven)
1529
def merge(self, branch_from, wt_to):
1530
"""A helper for tests to do a ui-less merge.
1532
This should move to the main library when someone has time to integrate
1535
# minimal ui-less merge.
1536
wt_to.branch.fetch(branch_from)
1537
base_rev = common_ancestor(branch_from.last_revision(),
1538
wt_to.branch.last_revision(),
1539
wt_to.branch.repository)
1540
merge_inner(wt_to.branch, branch_from.basis_tree(),
1541
wt_to.branch.repository.revision_tree(base_rev),
1543
wt_to.add_parent_tree_id(branch_from.last_revision())
1545
def reduceLockdirTimeout(self):
1546
"""Reduce the default lock timeout for the duration of the test, so that
1547
if LockContention occurs during a test, it does so quickly.
1549
Tests that expect to provoke LockContention errors should call this.
1551
orig_timeout = bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS
1553
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = orig_timeout
1554
self.addCleanup(resetTimeout)
1555
bzrlib.lockdir._DEFAULT_TIMEOUT_SECONDS = 0
1557
BzrTestBase = TestCase
1560
class TestCaseWithMemoryTransport(TestCase):
1561
"""Common test class for tests that do not need disk resources.
1563
Tests that need disk resources should derive from TestCaseWithTransport.
1565
TestCaseWithMemoryTransport sets the TEST_ROOT variable for all bzr tests.
1567
For TestCaseWithMemoryTransport the test_home_dir is set to the name of
1568
a directory which does not exist. This serves to help ensure test isolation
1569
is preserved. test_dir is set to the TEST_ROOT, as is cwd, because they
1570
must exist. However, TestCaseWithMemoryTransport does not offer local
1571
file defaults for the transport in tests, nor does it obey the command line
1572
override, so tests that accidentally write to the common directory should
1580
def __init__(self, methodName='runTest'):
1581
# allow test parameterisation after test construction and before test
1582
# execution. Variables that the parameteriser sets need to be
1583
# ones that are not set by setUp, or setUp will trash them.
1584
super(TestCaseWithMemoryTransport, self).__init__(methodName)
1585
self.vfs_transport_factory = default_transport
1586
self.transport_server = None
1587
self.transport_readonly_server = None
1588
self.__vfs_server = None
1590
def get_transport(self):
1591
"""Return a writeable transport for the test scratch space"""
1592
t = get_transport(self.get_url())
1593
self.assertFalse(t.is_readonly())
1596
def get_readonly_transport(self):
1597
"""Return a readonly transport for the test scratch space
1599
This can be used to test that operations which should only need
1600
readonly access in fact do not try to write.
1602
t = get_transport(self.get_readonly_url())
1603
self.assertTrue(t.is_readonly())
1606
def create_transport_readonly_server(self):
1607
"""Create a transport server from class defined at init.
1609
This is mostly a hook for daughter classes.
1611
return self.transport_readonly_server()
1613
def get_readonly_server(self):
1614
"""Get the server instance for the readonly transport
1616
This is useful for some tests with specific servers to do diagnostics.
1618
if self.__readonly_server is None:
1619
if self.transport_readonly_server is None:
1620
# readonly decorator requested
1621
# bring up the server
1622
self.__readonly_server = ReadonlyServer()
1623
self.__readonly_server.setUp(self.get_vfs_only_server())
1625
self.__readonly_server = self.create_transport_readonly_server()
1626
self.__readonly_server.setUp(self.get_vfs_only_server())
1627
self.addCleanup(self.__readonly_server.tearDown)
1628
return self.__readonly_server
1630
def get_readonly_url(self, relpath=None):
1631
"""Get a URL for the readonly transport.
1633
This will either be backed by '.' or a decorator to the transport
1634
used by self.get_url()
1635
relpath provides for clients to get a path relative to the base url.
1636
These should only be downwards relative, not upwards.
1638
base = self.get_readonly_server().get_url()
1639
if relpath is not None:
1640
if not base.endswith('/'):
1642
base = base + relpath
1645
def get_vfs_only_server(self):
1646
"""Get the vfs only read/write server instance.
1648
This is useful for some tests with specific servers that need
1651
For TestCaseWithMemoryTransport this is always a MemoryServer, and there
1652
is no means to override it.
1654
if self.__vfs_server is None:
1655
self.__vfs_server = MemoryServer()
1656
self.__vfs_server.setUp()
1657
self.addCleanup(self.__vfs_server.tearDown)
1658
return self.__vfs_server
1660
def get_server(self):
1661
"""Get the read/write server instance.
1663
This is useful for some tests with specific servers that need
1666
This is built from the self.transport_server factory. If that is None,
1667
then the self.get_vfs_server is returned.
1669
if self.__server is None:
1670
if self.transport_server is None or self.transport_server is self.vfs_transport_factory:
1671
return self.get_vfs_only_server()
1673
# bring up a decorated means of access to the vfs only server.
1674
self.__server = self.transport_server()
1676
self.__server.setUp(self.get_vfs_only_server())
1677
except TypeError, e:
1678
# This should never happen; the try:Except here is to assist
1679
# developers having to update code rather than seeing an
1680
# uninformative TypeError.
1681
raise Exception, "Old server API in use: %s, %s" % (self.__server, e)
1682
self.addCleanup(self.__server.tearDown)
1683
return self.__server
1685
def _adjust_url(self, base, relpath):
1686
"""Get a URL (or maybe a path) for the readwrite transport.
1688
This will either be backed by '.' or to an equivalent non-file based
1690
relpath provides for clients to get a path relative to the base url.
1691
These should only be downwards relative, not upwards.
1693
if relpath is not None and relpath != '.':
1694
if not base.endswith('/'):
1696
# XXX: Really base should be a url; we did after all call
1697
# get_url()! But sometimes it's just a path (from
1698
# LocalAbspathServer), and it'd be wrong to append urlescaped data
1699
# to a non-escaped local path.
1700
if base.startswith('./') or base.startswith('/'):
1703
base += urlutils.escape(relpath)
1706
def get_url(self, relpath=None):
1707
"""Get a URL (or maybe a path) for the readwrite transport.
1709
This will either be backed by '.' or to an equivalent non-file based
1711
relpath provides for clients to get a path relative to the base url.
1712
These should only be downwards relative, not upwards.
1714
base = self.get_server().get_url()
1715
return self._adjust_url(base, relpath)
1717
def get_vfs_only_url(self, relpath=None):
1718
"""Get a URL (or maybe a path for the plain old vfs transport.
1720
This will never be a smart protocol.
1721
:param relpath: provides for clients to get a path relative to the base
1722
url. These should only be downwards relative, not upwards.
1724
base = self.get_vfs_only_server().get_url()
1725
return self._adjust_url(base, relpath)
1727
def _make_test_root(self):
1728
if TestCaseWithMemoryTransport.TEST_ROOT is not None:
1732
root = u'test%04d.tmp' % i
1736
if e.errno == errno.EEXIST:
1741
# successfully created
1742
TestCaseWithMemoryTransport.TEST_ROOT = osutils.abspath(root)
1744
# make a fake bzr directory there to prevent any tests propagating
1745
# up onto the source directory's real branch
1746
bzrdir.BzrDir.create_standalone_workingtree(
1747
TestCaseWithMemoryTransport.TEST_ROOT)
1749
def makeAndChdirToTestDir(self):
1750
"""Create a temporary directories for this one test.
1752
This must set self.test_home_dir and self.test_dir and chdir to
1755
For TestCaseWithMemoryTransport we chdir to the TEST_ROOT for this test.
1757
os.chdir(TestCaseWithMemoryTransport.TEST_ROOT)
1758
self.test_dir = TestCaseWithMemoryTransport.TEST_ROOT
1759
self.test_home_dir = self.test_dir + "/MemoryTransportMissingHomeDir"
1761
def make_branch(self, relpath, format=None):
1762
"""Create a branch on the transport at relpath."""
1763
repo = self.make_repository(relpath, format=format)
1764
return repo.bzrdir.create_branch()
1766
def make_bzrdir(self, relpath, format=None):
1768
# might be a relative or absolute path
1769
maybe_a_url = self.get_url(relpath)
1770
segments = maybe_a_url.rsplit('/', 1)
1771
t = get_transport(maybe_a_url)
1772
if len(segments) > 1 and segments[-1] not in ('', '.'):
1775
except errors.FileExists:
1779
if isinstance(format, basestring):
1780
format = bzrdir.format_registry.make_bzrdir(format)
1781
return format.initialize_on_transport(t)
1782
except errors.UninitializableFormat:
1783
raise TestSkipped("Format %s is not initializable." % format)
1785
def make_repository(self, relpath, shared=False, format=None):
1786
"""Create a repository on our default transport at relpath."""
1787
made_control = self.make_bzrdir(relpath, format=format)
1788
return made_control.create_repository(shared=shared)
1790
def make_branch_and_memory_tree(self, relpath, format=None):
1791
"""Create a branch on the default transport and a MemoryTree for it."""
1792
b = self.make_branch(relpath, format=format)
1793
return memorytree.MemoryTree.create_on_branch(b)
1795
def overrideEnvironmentForTesting(self):
1796
os.environ['HOME'] = self.test_home_dir
1797
os.environ['BZR_HOME'] = self.test_home_dir
1800
super(TestCaseWithMemoryTransport, self).setUp()
1801
self._make_test_root()
1802
_currentdir = os.getcwdu()
1803
def _leaveDirectory():
1804
os.chdir(_currentdir)
1805
self.addCleanup(_leaveDirectory)
1806
self.makeAndChdirToTestDir()
1807
self.overrideEnvironmentForTesting()
1808
self.__readonly_server = None
1809
self.__server = None
1810
self.reduceLockdirTimeout()
1813
class TestCaseInTempDir(TestCaseWithMemoryTransport):
1814
"""Derived class that runs a test within a temporary directory.
1816
This is useful for tests that need to create a branch, etc.
1818
The directory is created in a slightly complex way: for each
1819
Python invocation, a new temporary top-level directory is created.
1820
All test cases create their own directory within that. If the
1821
tests complete successfully, the directory is removed.
1823
InTempDir is an old alias for FunctionalTestCase.
1826
OVERRIDE_PYTHON = 'python'
1828
def check_file_contents(self, filename, expect):
1829
self.log("check contents of file %s" % filename)
1830
contents = file(filename, 'r').read()
1831
if contents != expect:
1832
self.log("expected: %r" % expect)
1833
self.log("actually: %r" % contents)
1834
self.fail("contents of %s not as expected" % filename)
1836
def makeAndChdirToTestDir(self):
1837
"""See TestCaseWithMemoryTransport.makeAndChdirToTestDir().
1839
For TestCaseInTempDir we create a temporary directory based on the test
1840
name and then create two subdirs - test and home under it.
1842
if NUMBERED_DIRS: # strongly recommended on Windows
1843
# due the path length limitation (260 chars)
1844
candidate_dir = '%s/%dK/%05d' % (self.TEST_ROOT,
1845
int(self.number/1000),
1847
os.makedirs(candidate_dir)
1848
self.test_home_dir = candidate_dir + '/home'
1849
os.mkdir(self.test_home_dir)
1850
self.test_dir = candidate_dir + '/work'
1851
os.mkdir(self.test_dir)
1852
os.chdir(self.test_dir)
1853
# put name of test inside
1854
f = file(candidate_dir + '/name', 'w')
1859
# shorten the name, to avoid test failures due to path length
1860
short_id = self.id().replace('bzrlib.tests.', '') \
1861
.replace('__main__.', '')[-100:]
1862
# it's possible the same test class is run several times for
1863
# parameterized tests, so make sure the names don't collide.
1867
candidate_dir = '%s/%s.%d' % (self.TEST_ROOT, short_id, i)
1869
candidate_dir = '%s/%s' % (self.TEST_ROOT, short_id)
1870
if os.path.exists(candidate_dir):
1874
os.mkdir(candidate_dir)
1875
self.test_home_dir = candidate_dir + '/home'
1876
os.mkdir(self.test_home_dir)
1877
self.test_dir = candidate_dir + '/work'
1878
os.mkdir(self.test_dir)
1879
os.chdir(self.test_dir)
1882
def build_tree(self, shape, line_endings='binary', transport=None):
1883
"""Build a test tree according to a pattern.
1885
shape is a sequence of file specifications. If the final
1886
character is '/', a directory is created.
1888
This assumes that all the elements in the tree being built are new.
1890
This doesn't add anything to a branch.
1891
:param line_endings: Either 'binary' or 'native'
1892
in binary mode, exact contents are written
1893
in native mode, the line endings match the
1894
default platform endings.
1896
:param transport: A transport to write to, for building trees on
1897
VFS's. If the transport is readonly or None,
1898
"." is opened automatically.
1900
# It's OK to just create them using forward slashes on windows.
1901
if transport is None or transport.is_readonly():
1902
transport = get_transport(".")
1904
self.assert_(isinstance(name, basestring))
1906
transport.mkdir(urlutils.escape(name[:-1]))
1908
if line_endings == 'binary':
1910
elif line_endings == 'native':
1913
raise errors.BzrError(
1914
'Invalid line ending request %r' % line_endings)
1915
content = "contents of %s%s" % (name.encode('utf-8'), end)
1916
transport.put_bytes_non_atomic(urlutils.escape(name), content)
1918
def build_tree_contents(self, shape):
1919
build_tree_contents(shape)
1921
def assertFileEqual(self, content, path):
1922
"""Fail if path does not contain 'content'."""
1923
self.failUnlessExists(path)
1924
# TODO: jam 20060427 Shouldn't this be 'rb'?
1930
self.assertEqualDiff(content, s)
1932
def failUnlessExists(self, path):
1933
"""Fail unless path, which may be abs or relative, exists."""
1934
self.failUnless(osutils.lexists(path),path+" does not exist")
1936
def failIfExists(self, path):
1937
"""Fail if path, which may be abs or relative, exists."""
1938
self.failIf(osutils.lexists(path),path+" exists")
1941
class TestCaseWithTransport(TestCaseInTempDir):
1942
"""A test case that provides get_url and get_readonly_url facilities.
1944
These back onto two transport servers, one for readonly access and one for
1947
If no explicit class is provided for readonly access, a
1948
ReadonlyTransportDecorator is used instead which allows the use of non disk
1949
based read write transports.
1951
If an explicit class is provided for readonly access, that server and the
1952
readwrite one must both define get_url() as resolving to os.getcwd().
1955
def get_vfs_only_server(self):
1956
"""See TestCaseWithMemoryTransport.
1958
This is useful for some tests with specific servers that need
1961
if self.__vfs_server is None:
1962
self.__vfs_server = self.vfs_transport_factory()
1963
self.__vfs_server.setUp()
1964
self.addCleanup(self.__vfs_server.tearDown)
1965
return self.__vfs_server
1967
def make_branch_and_tree(self, relpath, format=None):
1968
"""Create a branch on the transport and a tree locally.
1970
If the transport is not a LocalTransport, the Tree can't be created on
1971
the transport. In that case if the vfs_transport_factory is
1972
LocalURLServer the working tree is created in the local
1973
directory backing the transport, and the returned tree's branch and
1974
repository will also be accessed locally. Otherwise a lightweight
1975
checkout is created and returned.
1977
:param format: The BzrDirFormat.
1978
:returns: the WorkingTree.
1980
# TODO: always use the local disk path for the working tree,
1981
# this obviously requires a format that supports branch references
1982
# so check for that by checking bzrdir.BzrDirFormat.get_default_format()
1984
b = self.make_branch(relpath, format=format)
1986
return b.bzrdir.create_workingtree()
1987
except errors.NotLocalUrl:
1988
# We can only make working trees locally at the moment. If the
1989
# transport can't support them, then we keep the non-disk-backed
1990
# branch and create a local checkout.
1991
if self.vfs_transport_factory is LocalURLServer:
1992
# the branch is colocated on disk, we cannot create a checkout.
1993
# hopefully callers will expect this.
1994
local_controldir= bzrdir.BzrDir.open(self.get_vfs_only_url(relpath))
1995
return local_controldir.create_workingtree()
1997
return b.create_checkout(relpath, lightweight=True)
1999
def assertIsDirectory(self, relpath, transport):
2000
"""Assert that relpath within transport is a directory.
2002
This may not be possible on all transports; in that case it propagates
2003
a TransportNotPossible.
2006
mode = transport.stat(relpath).st_mode
2007
except errors.NoSuchFile:
2008
self.fail("path %s is not a directory; no such file"
2010
if not stat.S_ISDIR(mode):
2011
self.fail("path %s is not a directory; has mode %#o"
2014
def assertTreesEqual(self, left, right):
2015
"""Check that left and right have the same content and properties."""
2016
# we use a tree delta to check for equality of the content, and we
2017
# manually check for equality of other things such as the parents list.
2018
self.assertEqual(left.get_parent_ids(), right.get_parent_ids())
2019
differences = left.changes_from(right)
2020
self.assertFalse(differences.has_changed(),
2021
"Trees %r and %r are different: %r" % (left, right, differences))
2024
super(TestCaseWithTransport, self).setUp()
2025
self.__vfs_server = None
2028
class ChrootedTestCase(TestCaseWithTransport):
2029
"""A support class that provides readonly urls outside the local namespace.
2031
This is done by checking if self.transport_server is a MemoryServer. if it
2032
is then we are chrooted already, if it is not then an HttpServer is used
2035
TODO RBC 20060127: make this an option to TestCaseWithTransport so it can
2036
be used without needed to redo it when a different
2037
subclass is in use ?
2041
super(ChrootedTestCase, self).setUp()
2042
if not self.vfs_transport_factory == MemoryServer:
2043
self.transport_readonly_server = HttpServer
2046
def filter_suite_by_re(suite, pattern):
2047
result = TestUtil.TestSuite()
2048
filter_re = re.compile(pattern)
2049
for test in iter_suite_tests(suite):
2050
if filter_re.search(test.id()):
2051
result.addTest(test)
2055
def sort_suite_by_re(suite, pattern):
2058
filter_re = re.compile(pattern)
2059
for test in iter_suite_tests(suite):
2060
if filter_re.search(test.id()):
2064
return TestUtil.TestSuite(first + second)
2067
def run_suite(suite, name='test', verbose=False, pattern=".*",
2068
stop_on_failure=False, keep_output=False,
2069
transport=None, lsprof_timed=None, bench_history=None,
2070
matching_tests_first=None,
2071
numbered_dirs=None):
2072
global NUMBERED_DIRS
2073
if numbered_dirs is not None:
2074
NUMBERED_DIRS = bool(numbered_dirs)
2076
TestCase._gather_lsprof_in_benchmarks = lsprof_timed
2081
runner = TextTestRunner(stream=sys.stdout,
2083
verbosity=verbosity,
2084
keep_output=keep_output,
2085
bench_history=bench_history)
2086
runner.stop_on_failure=stop_on_failure
2088
if matching_tests_first:
2089
suite = sort_suite_by_re(suite, pattern)
2091
suite = filter_suite_by_re(suite, pattern)
2092
result = runner.run(suite)
2093
return result.wasSuccessful()
2096
def selftest(verbose=False, pattern=".*", stop_on_failure=True,
2099
test_suite_factory=None,
2102
matching_tests_first=None,
2103
numbered_dirs=None):
2104
"""Run the whole test suite under the enhanced runner"""
2105
# XXX: Very ugly way to do this...
2106
# Disable warning about old formats because we don't want it to disturb
2107
# any blackbox tests.
2108
from bzrlib import repository
2109
repository._deprecation_warning_done = True
2111
global default_transport
2112
if transport is None:
2113
transport = default_transport
2114
old_transport = default_transport
2115
default_transport = transport
2117
if test_suite_factory is None:
2118
suite = test_suite()
2120
suite = test_suite_factory()
2121
return run_suite(suite, 'testbzr', verbose=verbose, pattern=pattern,
2122
stop_on_failure=stop_on_failure, keep_output=keep_output,
2123
transport=transport,
2124
lsprof_timed=lsprof_timed,
2125
bench_history=bench_history,
2126
matching_tests_first=matching_tests_first,
2127
numbered_dirs=numbered_dirs)
2129
default_transport = old_transport
2133
"""Build and return TestSuite for the whole of bzrlib.
2135
This function can be replaced if you need to change the default test
2136
suite on a global basis, but it is not encouraged.
2139
'bzrlib.tests.test_ancestry',
2140
'bzrlib.tests.test_annotate',
2141
'bzrlib.tests.test_api',
2142
'bzrlib.tests.test_atomicfile',
2143
'bzrlib.tests.test_bad_files',
2144
'bzrlib.tests.test_branch',
2145
'bzrlib.tests.test_bundle',
2146
'bzrlib.tests.test_bzrdir',
2147
'bzrlib.tests.test_cache_utf8',
2148
'bzrlib.tests.test_commands',
2149
'bzrlib.tests.test_commit',
2150
'bzrlib.tests.test_commit_merge',
2151
'bzrlib.tests.test_config',
2152
'bzrlib.tests.test_conflicts',
2153
'bzrlib.tests.test_decorators',
2154
'bzrlib.tests.test_delta',
2155
'bzrlib.tests.test_diff',
2156
'bzrlib.tests.test_dirstate',
2157
'bzrlib.tests.test_doc_generate',
2158
'bzrlib.tests.test_errors',
2159
'bzrlib.tests.test_escaped_store',
2160
'bzrlib.tests.test_extract',
2161
'bzrlib.tests.test_fetch',
2162
'bzrlib.tests.test_ftp_transport',
2163
'bzrlib.tests.test_generate_docs',
2164
'bzrlib.tests.test_generate_ids',
2165
'bzrlib.tests.test_globbing',
2166
'bzrlib.tests.test_gpg',
2167
'bzrlib.tests.test_graph',
2168
'bzrlib.tests.test_hashcache',
2169
'bzrlib.tests.test_http',
2170
'bzrlib.tests.test_http_response',
2171
'bzrlib.tests.test_https_ca_bundle',
2172
'bzrlib.tests.test_identitymap',
2173
'bzrlib.tests.test_ignores',
2174
'bzrlib.tests.test_inv',
2175
'bzrlib.tests.test_knit',
2176
'bzrlib.tests.test_lazy_import',
2177
'bzrlib.tests.test_lazy_regex',
2178
'bzrlib.tests.test_lockdir',
2179
'bzrlib.tests.test_lockable_files',
2180
'bzrlib.tests.test_log',
2181
'bzrlib.tests.test_memorytree',
2182
'bzrlib.tests.test_merge',
2183
'bzrlib.tests.test_merge3',
2184
'bzrlib.tests.test_merge_core',
2185
'bzrlib.tests.test_merge_directive',
2186
'bzrlib.tests.test_missing',
2187
'bzrlib.tests.test_msgeditor',
2188
'bzrlib.tests.test_nonascii',
2189
'bzrlib.tests.test_options',
2190
'bzrlib.tests.test_osutils',
2191
'bzrlib.tests.test_osutils_encodings',
2192
'bzrlib.tests.test_patch',
2193
'bzrlib.tests.test_patches',
2194
'bzrlib.tests.test_permissions',
2195
'bzrlib.tests.test_plugins',
2196
'bzrlib.tests.test_progress',
2197
'bzrlib.tests.test_reconcile',
2198
'bzrlib.tests.test_registry',
2199
'bzrlib.tests.test_repository',
2200
'bzrlib.tests.test_revert',
2201
'bzrlib.tests.test_revision',
2202
'bzrlib.tests.test_revisionnamespaces',
2203
'bzrlib.tests.test_revisiontree',
2204
'bzrlib.tests.test_rio',
2205
'bzrlib.tests.test_sampler',
2206
'bzrlib.tests.test_selftest',
2207
'bzrlib.tests.test_setup',
2208
'bzrlib.tests.test_sftp_transport',
2209
'bzrlib.tests.test_smart_add',
2210
'bzrlib.tests.test_smart_transport',
2211
'bzrlib.tests.test_source',
2212
'bzrlib.tests.test_ssh_transport',
2213
'bzrlib.tests.test_status',
2214
'bzrlib.tests.test_store',
2215
'bzrlib.tests.test_strace',
2216
'bzrlib.tests.test_subsume',
2217
'bzrlib.tests.test_symbol_versioning',
2218
'bzrlib.tests.test_tag',
2219
'bzrlib.tests.test_testament',
2220
'bzrlib.tests.test_textfile',
2221
'bzrlib.tests.test_textmerge',
2222
'bzrlib.tests.test_timestamp',
2223
'bzrlib.tests.test_trace',
2224
'bzrlib.tests.test_transactions',
2225
'bzrlib.tests.test_transform',
2226
'bzrlib.tests.test_transport',
2227
'bzrlib.tests.test_tree',
2228
'bzrlib.tests.test_treebuilder',
2229
'bzrlib.tests.test_tsort',
2230
'bzrlib.tests.test_tuned_gzip',
2231
'bzrlib.tests.test_ui',
2232
'bzrlib.tests.test_upgrade',
2233
'bzrlib.tests.test_urlutils',
2234
'bzrlib.tests.test_versionedfile',
2235
'bzrlib.tests.test_version',
2236
'bzrlib.tests.test_version_info',
2237
'bzrlib.tests.test_weave',
2238
'bzrlib.tests.test_whitebox',
2239
'bzrlib.tests.test_workingtree',
2240
'bzrlib.tests.test_workingtree_4',
2241
'bzrlib.tests.test_wsgi',
2242
'bzrlib.tests.test_xml',
2244
test_transport_implementations = [
2245
'bzrlib.tests.test_transport_implementations',
2246
'bzrlib.tests.test_read_bundle',
2248
suite = TestUtil.TestSuite()
2249
loader = TestUtil.TestLoader()
2250
suite.addTest(loader.loadTestsFromModuleNames(testmod_names))
2251
from bzrlib.transport import TransportTestProviderAdapter
2252
adapter = TransportTestProviderAdapter()
2253
adapt_modules(test_transport_implementations, adapter, loader, suite)
2254
for package in packages_to_test():
2255
suite.addTest(package.test_suite())
2256
for m in MODULES_TO_TEST:
2257
suite.addTest(loader.loadTestsFromModule(m))
2258
for m in MODULES_TO_DOCTEST:
2260
suite.addTest(doctest.DocTestSuite(m))
2261
except ValueError, e:
2262
print '**failed to get doctest for: %s\n%s' %(m,e)
2264
for name, plugin in bzrlib.plugin.all_plugins().items():
2265
if getattr(plugin, 'test_suite', None) is not None:
2266
default_encoding = sys.getdefaultencoding()
2268
plugin_suite = plugin.test_suite()
2269
except ImportError, e:
2270
bzrlib.trace.warning(
2271
'Unable to test plugin "%s": %s', name, e)
2273
suite.addTest(plugin_suite)
2274
if default_encoding != sys.getdefaultencoding():
2275
bzrlib.trace.warning(
2276
'Plugin "%s" tried to reset default encoding to: %s', name,
2277
sys.getdefaultencoding())
2279
sys.setdefaultencoding(default_encoding)
2283
def adapt_modules(mods_list, adapter, loader, suite):
2284
"""Adapt the modules in mods_list using adapter and add to suite."""
2285
for test in iter_suite_tests(loader.loadTestsFromModuleNames(mods_list)):
2286
suite.addTests(adapter.adapt(test))
2289
def clean_selftest_output(root=None, quiet=False):
2290
"""Remove all selftest output directories from root directory.
2292
:param root: root directory for clean
2293
(if ommitted or None then clean current directory).
2294
:param quiet: suppress report about deleting directories
2299
re_dir = re.compile(r'''test\d\d\d\d\.tmp''')
2302
for i in os.listdir(root):
2303
if os.path.isdir(i) and re_dir.match(i):
2305
print 'delete directory:', i
2309
class Feature(object):
2310
"""An operating system Feature."""
2313
self._available = None
2315
def available(self):
2316
"""Is the feature available?
2318
:return: True if the feature is available.
2320
if self._available is None:
2321
self._available = self._probe()
2322
return self._available
2325
"""Implement this method in concrete features.
2327
:return: True if the feature is available.
2329
raise NotImplementedError
2332
if getattr(self, 'feature_name', None):
2333
return self.feature_name()
2334
return self.__class__.__name__
added
removed
bzrlib/tests/__init__.py
