6
The Importance of Testing
7
=========================
9
Reliability is a critical success factor for any Version Control System.
10
We want Bazaar to be highly reliable across multiple platforms while
11
evolving over time to meet the needs of its community.
13
In a nutshell, this is what we expect and encourage:
15
* New functionality should have test cases. Preferably write the
16
test before writing the code.
18
In general, you can test at either the command-line level or the
19
internal API level. See `Writing tests`_ below for more detail.
21
* Try to practice Test-Driven Development: before fixing a bug, write a
22
test case so that it does not regress. Similarly for adding a new
23
feature: write a test case for a small version of the new feature before
24
starting on the code itself. Check the test fails on the old code, then
25
add the feature or fix and check it passes.
27
By doing these things, the Bazaar team gets increased confidence that
28
changes do what they claim to do, whether provided by the core team or
29
by community members. Equally importantly, we can be surer that changes
30
down the track do not break new features or bug fixes that you are
33
As of September 2009, Bazaar ships with a test suite containing over
34
23,000 tests and growing. We are proud of it and want to remain so. As
35
community members, we all benefit from it. Would you trust version control
36
on your project to a product *without* a test suite like Bazaar has?
39
Running the Test Suite
40
======================
42
As of Bazaar 2.1, you must have the testtools_ library installed to run
45
.. _testtools: https://launchpad.net/testtools/
47
To test all of Bazaar, just run::
51
With ``--verbose`` bzr will print the name of every test as it is run.
53
This should always pass, whether run from a source tree or an installed
54
copy of Bazaar. Please investigate and/or report any failures.
57
Running particular tests
58
------------------------
60
Currently, bzr selftest is used to invoke tests.
61
You can provide a pattern argument to run a subset. For example,
62
to run just the blackbox tests, run::
64
./bzr selftest -v blackbox
66
To skip a particular test (or set of tests), use the --exclude option
67
(shorthand -x) like so::
69
./bzr selftest -v -x blackbox
71
To ensure that all tests are being run and succeeding, you can use the
72
--strict option which will fail if there are any missing features or known
75
./bzr selftest --strict
77
To list tests without running them, use the --list-only option like so::
79
./bzr selftest --list-only
81
This option can be combined with other selftest options (like -x) and
82
filter patterns to understand their effect.
84
Once you understand how to create a list of tests, you can use the --load-list
85
option to run only a restricted set of tests that you kept in a file, one test
86
id by line. Keep in mind that this will never be sufficient to validate your
87
modifications, you still need to run the full test suite for that, but using it
88
can help in some cases (like running only the failed tests for some time)::
90
./bzr selftest -- load-list my_failing_tests
92
This option can also be combined with other selftest options, including
93
patterns. It has some drawbacks though, the list can become out of date pretty
94
quick when doing Test Driven Development.
96
To address this concern, there is another way to run a restricted set of tests:
97
the --starting-with option will run only the tests whose name starts with the
98
specified string. It will also avoid loading the other tests and as a
99
consequence starts running your tests quicker::
101
./bzr selftest --starting-with bzrlib.blackbox
103
This option can be combined with all the other selftest options including
104
--load-list. The later is rarely used but allows to run a subset of a list of
105
failing tests for example.
110
To test only the bzr core, ignoring any plugins you may have installed,
113
./bzr --no-plugins selftest
115
Disabling crash reporting
116
-------------------------
118
By default Bazaar uses apport_ to report program crashes. In developing
119
Bazaar it's normal and expected to have it crash from time to time, at
120
least because a test failed if for no other reason.
122
Therefore you should probably add ``debug_flags = no_apport`` to your
123
``bazaar.conf`` file (in ``~/.bazaar/`` on Unix), so that failures just
124
print a traceback rather than writing a crash file.
126
.. _apport: https://launchpad.net/apport/
129
Test suite debug flags
130
----------------------
132
Similar to the global ``-Dfoo`` debug options, bzr selftest accepts
133
``-E=foo`` debug flags. These flags are:
135
:allow_debug: do *not* clear the global debug flags when running a test.
136
This can provide useful logging to help debug test failures when used
137
with e.g. ``bzr -Dhpss selftest -E=allow_debug``
139
Note that this will probably cause some tests to fail, because they
140
don't expect to run with any debug flags on.
146
Bazaar can optionally produce output in the machine-readable subunit_
147
format, so that test output can be post-processed by various tools. To
148
generate a subunit test stream::
150
$ ./bzr selftest --subunit
152
Processing such a stream can be done using a variety of tools including:
154
* The builtin ``subunit2pyunit``, ``subunit-filter``, ``subunit-ls``,
155
``subunit2junitxml`` from the subunit project.
157
* tribunal_, a GUI for showing test results.
159
* testrepository_, a tool for gathering and managing test runs.
161
.. _subunit: https://launchpad.net/subunit/
162
.. _tribunal: https://launchpad.net/tribunal/
168
Bazaar ships with a config file for testrepository_. This can be very
169
useful for keeping track of failing tests and doing general workflow
170
support. To run tests using testrepository::
174
To run only failing tests::
176
$ testr run --failing
178
To run only some tests, without plugins::
180
$ test run test_selftest -- --no-plugins
182
See the testrepository documentation for more details.
184
.. _testrepository: https://launchpad.net/testrepository
189
Normally you should add or update a test for all bug fixes or new features
193
Where should I put a new test?
194
------------------------------
196
Bzrlib's tests are organised by the type of test. Most of the tests in
197
bzr's test suite belong to one of these categories:
200
- Blackbox (UI) tests
201
- Per-implementation tests
204
A quick description of these test types and where they belong in bzrlib's
205
source follows. Not all tests fall neatly into one of these categories;
206
in those cases use your judgement.
212
Unit tests make up the bulk of our test suite. These are tests that are
213
focused on exercising a single, specific unit of the code as directly
214
as possible. Each unit test is generally fairly short and runs very
217
They are found in ``bzrlib/tests/test_*.py``. So in general tests should
218
be placed in a file named test_FOO.py where FOO is the logical thing under
221
For example, tests for merge3 in bzrlib belong in bzrlib/tests/test_merge3.py.
222
See bzrlib/tests/test_sampler.py for a template test script.
228
Tests can be written for the UI or for individual areas of the library.
229
Choose whichever is appropriate: if adding a new command, or a new command
230
option, then you should be writing a UI test. If you are both adding UI
231
functionality and library functionality, you will want to write tests for
232
both the UI and the core behaviours. We call UI tests 'blackbox' tests
233
and they belong in ``bzrlib/tests/blackbox/*.py``.
235
When writing blackbox tests please honour the following conventions:
237
1. Place the tests for the command 'name' in
238
bzrlib/tests/blackbox/test_name.py. This makes it easy for developers
239
to locate the test script for a faulty command.
241
2. Use the 'self.run_bzr("name")' utility function to invoke the command
242
rather than running bzr in a subprocess or invoking the
243
cmd_object.run() method directly. This is a lot faster than
244
subprocesses and generates the same logging output as running it in a
245
subprocess (which invoking the method directly does not).
247
3. Only test the one command in a single test script. Use the bzrlib
248
library when setting up tests and when evaluating the side-effects of
249
the command. We do this so that the library api has continual pressure
250
on it to be as functional as the command line in a simple manner, and
251
to isolate knock-on effects throughout the blackbox test suite when a
252
command changes its name or signature. Ideally only the tests for a
253
given command are affected when a given command is changed.
255
4. If you have a test which does actually require running bzr in a
256
subprocess you can use ``run_bzr_subprocess``. By default the spawned
257
process will not load plugins unless ``--allow-plugins`` is supplied.
260
Per-implementation tests
261
~~~~~~~~~~~~~~~~~~~~~~~~
263
Per-implementation tests are tests that are defined once and then run
264
against multiple implementations of an interface. For example,
265
``per_transport.py`` defines tests that all Transport implementations
266
(local filesystem, HTTP, and so on) must pass. They are found in
267
``bzrlib/tests/per_*/*.py``, and ``bzrlib/tests/per_*.py``.
269
These are really a sub-category of unit tests, but an important one.
271
Along the same lines are tests for extension modules. We generally have
272
both a pure-python and a compiled implementation for each module. As such,
273
we want to run the same tests against both implementations. These can
274
generally be found in ``bzrlib/tests/*__*.py`` since extension modules are
275
usually prefixed with an underscore. Since there are only two
276
implementations, we have a helper function
277
``bzrlib.tests.permute_for_extension``, which can simplify the
278
``load_tests`` implementation.
284
We make selective use of doctests__. In general they should provide
285
*examples* within the API documentation which can incidentally be tested. We
286
don't try to test every important case using doctests |--| regular Python
287
tests are generally a better solution. That is, we just use doctests to
288
make our documentation testable, rather than as a way to make tests.
290
Most of these are in ``bzrlib/doc/api``. More additions are welcome.
292
__ http://docs.python.org/lib/module-doctest.html
298
``bzrlib/tests/script.py`` allows users to write tests in a syntax very close to a shell session,
299
using a restricted and limited set of commands that should be enough to mimic
300
most of the behaviours.
302
A script is a set of commands, each command is composed of:
304
* one mandatory command line,
305
* one optional set of input lines to feed the command,
306
* one optional set of output expected lines,
307
* one optional set of error expected lines.
309
Input, output and error lines can be specified in any order.
311
Except for the expected output, all lines start with a special
312
string (based on their origin when used under a Unix shell):
314
* '$ ' for the command,
316
* nothing for output,
319
Comments can be added anywhere, they start with '#' and end with
322
The execution stops as soon as an expected output or an expected error is not
325
When no output is specified, any ouput from the command is accepted
326
and execution continue.
328
If an error occurs and no expected error is specified, the execution stops.
330
An error is defined by a returned status different from zero, not by the
331
presence of text on the error stream.
333
The matching is done on a full string comparison basis unless '...' is used, in
334
which case expected output/errors can be less precise.
338
The following will succeeds only if 'bzr add' outputs 'adding file'::
343
If you want the command to succeed for any output, just use::
347
The following will stop with an error::
351
If you want it to succeed, use::
354
2> bzr: ERROR: unknown command "not-a-command"
356
You can use ellipsis (...) to replace any piece of text you don't want to be
359
$ bzr branch not-a-branch
360
2>bzr: ERROR: Not a branch...not-a-branch/".
362
This can be used to ignore entire lines too::
368
# And here we explain that surprising fourth line
375
You can check the content of a file with cat::
380
You can also check the existence of a file with cat, the following will fail if
381
the file doesn't exist::
385
The actual use of ScriptRunner within a TestCase looks something like
388
def test_unshelve_keep(self):
391
sr.run_script(self, '''
393
$ bzr shelve --all -m Foo
396
$ bzr unshelve --keep
407
`bzrlib.tests.test_import_tariff` has some tests that measure how many
408
Python modules are loaded to run some representative commands.
410
We want to avoid loading code unnecessarily, for reasons including:
412
* Python modules are interpreted when they're loaded, either to define
413
classes or modules or perhaps to initialize some structures.
415
* With a cold cache we may incur blocking real disk IO for each module.
417
* Some modules depend on many others.
419
* Some optional modules such as `testtools` are meant to be soft
420
dependencies and only needed for particular cases. If they're loaded in
421
other cases then bzr may break for people who don't have those modules.
423
`test_import_tarrif` allows us to check that removal of imports doesn't
426
This is done by running the command in a subprocess with
427
``--profile-imports``. Starting a whole Python interpreter is pretty
428
slow, so we don't want exhaustive testing here, but just enough to guard
429
against distinct fixed problems.
431
Assertions about precisely what is loaded tend to be brittle so we instead
432
make assertions that particular things aren't loaded.
434
Unless selftest is run with ``--no-plugins``, modules will be loaded in
435
the usual way and checks made on what they cause to be loaded. This is
436
probably worth checking into, because many bzr users have at least some
437
plugins installed (and they're included in binary installers).
439
In theory, plugins might have a good reason to load almost anything:
440
someone might write a plugin that opens a network connection or pops up a
441
gui window every time you run 'bzr status'. However, it's more likely
442
that the code to do these things is just being loaded accidentally. We
443
might eventually need to have a way to make exceptions for particular
446
Some things to check:
448
* non-GUI commands shouldn't load GUI libraries
450
* operations on bzr native formats sholudn't load foreign branch libraries
452
* network code shouldn't be loaded for purely local operations
454
* particularly expensive Python built-in modules shouldn't be loaded
455
unless there is a good reason
458
Testing locking behaviour
459
-------------------------
461
In order to test the locking behaviour of commands, it is possible to install
462
a hook that is called when a write lock is: acquired, released or broken.
463
(Read locks also exist, they cannot be discovered in this way.)
465
A hook can be installed by calling bzrlib.lock.Lock.hooks.install_named_hook.
466
The three valid hooks are: `lock_acquired`, `lock_released` and `lock_broken`.
473
lock.Lock.hooks.install_named_hook('lock_acquired',
474
locks_acquired.append, None)
475
lock.Lock.hooks.install_named_hook('lock_released',
476
locks_released.append, None)
478
`locks_acquired` will now receive a LockResult instance for all locks acquired
479
since the time the hook is installed.
481
The last part of the `lock_url` allows you to identify the type of object that is locked.
483
- BzrDir: `/branch-lock`
484
- Working tree: `/checkout/lock`
485
- Branch: `/branch/lock`
486
- Repository: `/repository/lock`
488
To test if a lock is a write lock on a working tree, one can do the following::
490
self.assertEndsWith(locks_acquired[0].lock_url, "/checkout/lock")
492
See bzrlib/tests/commands/test_revert.py for an example of how to use this for
499
In our enhancements to unittest we allow for some addition results beyond
500
just success or failure.
502
If a test can't be run, it can say that it's skipped by raising a special
503
exception. This is typically used in parameterized tests |--| for example
504
if a transport doesn't support setting permissions, we'll skip the tests
505
that relating to that. ::
508
return self.branch_format.initialize(repo.bzrdir)
509
except errors.UninitializableFormat:
510
raise tests.TestSkipped('Uninitializable branch format')
512
Raising TestSkipped is a good idea when you want to make it clear that the
513
test was not run, rather than just returning which makes it look as if it
516
Several different cases are distinguished:
519
Generic skip; the only type that was present up to bzr 0.18.
522
The test doesn't apply to the parameters with which it was run.
523
This is typically used when the test is being applied to all
524
implementations of an interface, but some aspects of the interface
525
are optional and not present in particular concrete
526
implementations. (Some tests that should raise this currently
527
either silently return or raise TestSkipped.) Another option is
528
to use more precise parameterization to avoid generating the test
532
The test can't be run because a dependency (typically a Python
533
library) is not available in the test environment. These
534
are in general things that the person running the test could fix
535
by installing the library. It's OK if some of these occur when
536
an end user runs the tests or if we're specifically testing in a
537
limited environment, but a full test should never see them.
539
See `Test feature dependencies`_ below.
542
The test exists but is known to fail, for example this might be
543
appropriate to raise if you've committed a test for a bug but not
544
the fix for it, or if something works on Unix but not on Windows.
546
Raising this allows you to distinguish these failures from the
547
ones that are not expected to fail. If the test would fail
548
because of something we don't expect or intend to fix,
549
KnownFailure is not appropriate, and TestNotApplicable might be
552
KnownFailure should be used with care as we don't want a
553
proliferation of quietly broken tests.
557
We plan to support three modes for running the test suite to control the
558
interpretation of these results. Strict mode is for use in situations
559
like merges to the mainline and releases where we want to make sure that
560
everything that can be tested has been tested. Lax mode is for use by
561
developers who want to temporarily tolerate some known failures. The
562
default behaviour is obtained by ``bzr selftest`` with no options, and
563
also (if possible) by running under another unittest harness.
565
======================= ======= ======= ========
566
result strict default lax
567
======================= ======= ======= ========
568
TestSkipped pass pass pass
569
TestNotApplicable pass pass pass
570
UnavailableFeature fail pass pass
571
KnownFailure fail pass pass
572
======================= ======= ======= ========
575
Test feature dependencies
576
-------------------------
578
Writing tests that require a feature
579
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
581
Rather than manually checking the environment in each test, a test class
582
can declare its dependence on some test features. The feature objects are
583
checked only once for each run of the whole test suite.
585
(For historical reasons, as of May 2007 many cases that should depend on
586
features currently raise TestSkipped.)
590
class TestStrace(TestCaseWithTransport):
592
_test_needs_features = [StraceFeature]
594
This means all tests in this class need the feature. If the feature is
595
not available the test will be skipped using UnavailableFeature.
597
Individual tests can also require a feature using the ``requireFeature``
600
self.requireFeature(StraceFeature)
602
The old naming style for features is CamelCase, but because they're
603
actually instances not classses they're now given instance-style names
606
Features already defined in ``bzrlib.tests`` and ``bzrlib.tests.features``
614
- UnicodeFilenameFeature
616
- CaseInsensitiveFilesystemFeature.
617
- chown_feature: The test can rely on OS being POSIX and python
619
- posix_permissions_feature: The test can use POSIX-style
620
user/group/other permission bits.
623
Defining a new feature that tests can require
624
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
626
New features for use with ``_test_needs_features`` or ``requireFeature``
627
are defined by subclassing ``bzrlib.tests.Feature`` and overriding the
628
``_probe`` and ``feature_name`` methods. For example::
630
class _SymlinkFeature(Feature):
633
return osutils.has_symlinks()
635
def feature_name(self):
638
SymlinkFeature = _SymlinkFeature()
640
A helper for handling running tests based on whether a python
641
module is available. This can handle 3rd-party dependencies (is
642
``paramiko`` available?) as well as stdlib (``termios``) or
643
extension modules (``bzrlib._groupcompress_pyx``). You create a
644
new feature instance with::
646
# in bzrlib/tests/features.py
647
apport = tests.ModuleAvailableFeature('apport')
650
# then in bzrlib/tests/test_apport.py
651
class TestApportReporting(TestCaseInTempDir):
653
_test_needs_features = [features.apport]
656
Testing exceptions and errors
657
-----------------------------
659
It's important to test handling of errors and exceptions. Because this
660
code is often not hit in ad-hoc testing it can often have hidden bugs --
661
it's particularly common to get NameError because the exception code
662
references a variable that has since been renamed.
664
.. TODO: Something about how to provoke errors in the right way?
666
In general we want to test errors at two levels:
668
1. A test in ``test_errors.py`` checking that when the exception object is
669
constructed with known parameters it produces an expected string form.
670
This guards against mistakes in writing the format string, or in the
671
``str`` representations of its parameters. There should be one for
672
each exception class.
674
2. Tests that when an api is called in a particular situation, it raises
675
an error of the expected class. You should typically use
676
``assertRaises``, which in the Bazaar test suite returns the exception
677
object to allow you to examine its parameters.
679
In some cases blackbox tests will also want to check error reporting. But
680
it can be difficult to provoke every error through the commandline
681
interface, so those tests are only done as needed |--| eg in response to a
682
particular bug or if the error is reported in an unusual way(?) Blackbox
683
tests should mostly be testing how the command-line interface works, so
684
should only test errors if there is something particular to the cli in how
685
they're displayed or handled.
691
The Python ``warnings`` module is used to indicate a non-fatal code
692
problem. Code that's expected to raise a warning can be tested through
695
The test suite can be run with ``-Werror`` to check no unexpected errors
698
However, warnings should be used with discretion. It's not an appropriate
699
way to give messages to the user, because the warning is normally shown
700
only once per source line that causes the problem. You should also think
701
about whether the warning is serious enought that it should be visible to
702
users who may not be able to fix it.
705
Interface implementation testing and test scenarios
706
---------------------------------------------------
708
There are several cases in Bazaar of multiple implementations of a common
709
conceptual interface. ("Conceptual" because it's not necessary for all
710
the implementations to share a base class, though they often do.)
711
Examples include transports and the working tree, branch and repository
714
In these cases we want to make sure that every implementation correctly
715
fulfils the interface requirements. For example, every Transport should
716
support the ``has()`` and ``get()`` and ``clone()`` methods. We have a
717
sub-suite of tests in ``test_transport_implementations``. (Most
718
per-implementation tests are in submodules of ``bzrlib.tests``, but not
719
the transport tests at the moment.)
721
These tests are repeated for each registered Transport, by generating a
722
new TestCase instance for the cross product of test methods and transport
723
implementations. As each test runs, it has ``transport_class`` and
724
``transport_server`` set to the class it should test. Most tests don't
725
access these directly, but rather use ``self.get_transport`` which returns
726
a transport of the appropriate type.
728
The goal is to run per-implementation only the tests that relate to that
729
particular interface. Sometimes we discover a bug elsewhere that happens
730
with only one particular transport. Once it's isolated, we can consider
731
whether a test should be added for that particular implementation,
732
or for all implementations of the interface.
734
The multiplication of tests for different implementations is normally
735
accomplished by overriding the ``load_tests`` function used to load tests
736
from a module. This function typically loads all the tests, then applies
737
a TestProviderAdapter to them, which generates a longer suite containing
738
all the test variations.
740
See also `Per-implementation tests`_ (above).
746
Some utilities are provided for generating variations of tests. This can
747
be used for per-implementation tests, or other cases where the same test
748
code needs to run several times on different scenarios.
750
The general approach is to define a class that provides test methods,
751
which depend on attributes of the test object being pre-set with the
752
values to which the test should be applied. The test suite should then
753
also provide a list of scenarios in which to run the tests.
755
Typically ``multiply_tests_from_modules`` should be called from the test
756
module's ``load_tests`` function.
762
We have a rich collection of tools to support writing tests. Please use
763
them in preference to ad-hoc solutions as they provide portability and
764
performance benefits.
767
TestCase and its subclasses
768
~~~~~~~~~~~~~~~~~~~~~~~~~~~
770
The ``bzrlib.tests`` module defines many TestCase classes to help you
774
A base TestCase that extends the Python standard library's
775
TestCase in several ways. It adds more assertion methods (e.g.
776
``assertContainsRe``), ``addCleanup``, and other features (see its API
777
docs for details). It also has a ``setUp`` that makes sure that
778
global state like registered hooks and loggers won't interfere with
779
your test. All tests should use this base class (whether directly or
782
TestCaseWithMemoryTransport
783
Extends TestCase and adds methods like ``get_transport``,
784
``make_branch`` and ``make_branch_builder``. The files created are
785
stored in a MemoryTransport that is discarded at the end of the test.
786
This class is good for tests that need to make branches or use
787
transports, but that don't require storing things on disk. All tests
788
that create bzrdirs should use this base class (either directly or via
789
a subclass) as it ensures that the test won't accidentally operate on
790
real branches in your filesystem.
793
Extends TestCaseWithMemoryTransport. For tests that really do need
794
files to be stored on disk, e.g. because a subprocess uses a file, or
795
for testing functionality that accesses the filesystem directly rather
796
than via the Transport layer (such as dirstate).
798
TestCaseWithTransport
799
Extends TestCaseInTempDir. Provides ``get_url`` and
800
``get_readonly_url`` facilities. Subclasses can control the
801
transports used by setting ``vfs_transport_factory``,
802
``transport_server`` and/or ``transport_readonly_server``.
805
See the API docs for more details.
811
When writing a test for a feature, it is often necessary to set up a
812
branch with a certain history. The ``BranchBuilder`` interface allows the
813
creation of test branches in a quick and easy manner. Here's a sample
816
builder = self.make_branch_builder('relpath')
817
builder.build_commit()
818
builder.build_commit()
819
builder.build_commit()
820
branch = builder.get_branch()
822
``make_branch_builder`` is a method of ``TestCaseWithMemoryTransport``.
824
Note that many current tests create test branches by inheriting from
825
``TestCaseWithTransport`` and using the ``make_branch_and_tree`` helper to
826
give them a ``WorkingTree`` that they can commit to. However, using the
827
newer ``make_branch_builder`` helper is preferred, because it can build
828
the changes in memory, rather than on disk. Tests that are explictly
829
testing how we work with disk objects should, of course, use a real
832
Please see bzrlib.branchbuilder for more details.
834
If you're going to examine the commit timestamps e.g. in a test for log
835
output, you should set the timestamp on the tree, rather than using fuzzy
842
The ``TreeBuilder`` interface allows the construction of arbitrary trees
843
with a declarative interface. A sample session might look like::
845
tree = self.make_branch_and_tree('path')
846
builder = TreeBuilder()
847
builder.start_tree(tree)
848
builder.build(['foo', "bar/", "bar/file"])
849
tree.commit('commit the tree')
850
builder.finish_tree()
852
Usually a test will create a tree using ``make_branch_and_memory_tree`` (a
853
method of ``TestCaseWithMemoryTransport``) or ``make_branch_and_tree`` (a
854
method of ``TestCaseWithTransport``).
856
Please see bzrlib.treebuilder for more details.
859
.. |--| unicode:: U+2014
added
removed
doc/developers/testing.txt
