193
193
<http://babune.ladeuil.net:24842/>. (Babune = Bazaar Buildbot Network.)
196
Running tests in parallel
197
-------------------------
199
Bazaar can use subunit to spawn multiple test processes. There is
200
slightly more chance you will hit ordering or timing-dependent bugs but
203
$ ./bzr selftest --parallel=fork
205
Note that you will need the Subunit library
206
<https://launchpad.net/subunit/> to use this, which is in
207
``python-subunit`` on Ubuntu.
210
Running tests from a ramdisk
211
----------------------------
213
The tests create and delete a lot of temporary files. In some cases you
214
can make the test suite run much faster by running it on a ramdisk. For
218
$ sudo mount -t tmpfs none /ram
219
$ TMPDIR=/ram ./bzr selftest ...
221
You could also change ``/tmp`` in ``/etc/fstab`` to have type ``tmpfs``,
222
if you don't mind possibly losing other files in there when the machine
223
restarts. Add this line (if there is none for ``/tmp`` already)::
225
none /tmp tmpfs defaults 0 0
227
With a 6-core machine and ``--parallel=fork`` using a tmpfs doubles the
228
test execution speed.
329
294
We make selective use of doctests__. In general they should provide
330
295
*examples* within the API documentation which can incidentally be tested. We
331
296
don't try to test every important case using doctests |--| regular Python
332
tests are generally a better solution. That is, we just use doctests to make
333
our documentation testable, rather than as a way to make tests. Be aware that
334
doctests are not as well isolated as the unit tests, if you need more
335
isolation, you're likely want to write unit tests anyway if only to get a
336
better control of the test environment.
297
tests are generally a better solution. That is, we just use doctests to
298
make our documentation testable, rather than as a way to make tests.
338
300
Most of these are in ``bzrlib/doc/api``. More additions are welcome.
346
``bzrlib/tests/script.py`` allows users to write tests in a syntax very
347
close to a shell session, using a restricted and limited set of commands
348
that should be enough to mimic most of the behaviours.
308
``bzrlib/tests/script.py`` allows users to write tests in a syntax very close to a shell session,
309
using a restricted and limited set of commands that should be enough to mimic
310
most of the behaviours.
350
312
A script is a set of commands, each command is composed of:
370
332
The execution stops as soon as an expected output or an expected error is not
373
If output occurs and no output is expected, the execution stops and the
374
test fails. If unexpected output occurs on the standard error, then
375
execution stops and the test fails.
335
When no output is specified, any ouput from the command is accepted
336
and execution continue.
377
338
If an error occurs and no expected error is specified, the execution stops.
440
You can run files containing shell-like scripts with::
442
$ bzr test-script <script>
444
where ``<script>`` is the path to the file containing the shell-like script.
446
395
The actual use of ScriptRunner within a TestCase looks something like
449
from bzrlib.tests import script
451
def test_unshelve_keep(self):
453
script.run_script(self, '''
455
$ bzr shelve -q --all -m Foo
458
$ bzr unshelve -q --keep
465
You can also test commands that read user interaction::
467
def test_confirm_action(self):
468
"""You can write tests that demonstrate user confirmation"""
469
commands.builtin_command_registry.register(cmd_test_confirm)
470
self.addCleanup(commands.builtin_command_registry.remove, 'test-confirm')
473
2>Really do it? [y/n]:
478
To avoid having to specify "-q" for all commands whose output is
479
irrelevant, the run_script() method may be passed the keyword argument
480
``null_output_matches_anything=True``. For example::
482
def test_ignoring_null_output(self):
485
$ bzr ci -m 'first revision' --unchanged
488
""", null_output_matches_anything=True)
398
def test_unshelve_keep(self):
401
sr.run_script(self, '''
403
$ bzr shelve --all -m Foo
406
$ bzr unshelve --keep
491
414
Import tariff tests
492
415
-------------------
740
663
_test_needs_features = [features.apport]
743
Testing deprecated code
744
-----------------------
746
When code is deprecated, it is still supported for some length of time,
747
usually until the next major version. The ``applyDeprecated`` helper
748
wraps calls to deprecated code to verify that it is correctly issuing the
749
deprecation warning, and also prevents the warnings from being printed
752
Typically patches that apply the ``@deprecated_function`` decorator should
753
update the accompanying tests to use the ``applyDeprecated`` wrapper.
755
``applyDeprecated`` is defined in ``bzrlib.tests.TestCase``. See the API
756
docs for more details.
759
666
Testing exceptions and errors
760
667
-----------------------------
834
741
whether a test should be added for that particular implementation,
835
742
or for all implementations of the interface.
744
The multiplication of tests for different implementations is normally
745
accomplished by overriding the ``load_tests`` function used to load tests
746
from a module. This function typically loads all the tests, then applies
747
a TestProviderAdapter to them, which generates a longer suite containing
748
all the test variations.
837
750
See also `Per-implementation tests`_ (above).
840
Test scenarios and variations
841
-----------------------------
843
756
Some utilities are provided for generating variations of tests. This can
844
757
be used for per-implementation tests, or other cases where the same test
849
762
values to which the test should be applied. The test suite should then
850
763
also provide a list of scenarios in which to run the tests.
852
A single *scenario* is defined by a `(name, parameter_dict)` tuple. The
853
short string name is combined with the name of the test method to form the
854
test instance name. The parameter dict is merged into the instance's
859
load_tests = load_tests_apply_scenarios
861
class TestCheckout(TestCase):
863
scenarios = multiply_scenarios(
864
VaryByRepositoryFormat(),
868
The `load_tests` declaration or definition should be near the top of the
869
file so its effect can be seen.
765
Typically ``multiply_tests_from_modules`` should be called from the test
766
module's ``load_tests`` function.
972
869
Please see bzrlib.treebuilder for more details.
975
Temporarily changing state
976
~~~~~~~~~~~~~~~~~~~~~~~~~~
978
If your test needs to temporarily mutate some global state, and you need
979
it restored at the end, you can say for example::
981
self.overrideAttr(osutils, '_cached_user_encoding', 'latin-1')
983
Temporarily changing environment variables
984
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
986
If yout test needs to temporarily change some environment variable value
987
(which generally means you want it restored at the end), you can use::
989
self.overrideEnv('BZR_ENV_VAR', 'new_value')
991
If you want to remove a variable from the environment, you should use the
992
special ``None`` value::
994
self.overrideEnv('PATH', None)
996
If you add a new feature which depends on a new environment variable, make
997
sure it behaves properly when this variable is not defined (if applicable) and
998
if you need to enforce a specific default value, check the
999
``TestCase._cleanEnvironment`` in ``bzrlib.tests.__init__.py`` which defines a
1000
proper set of values for all tests.
1005
Our base ``TestCase`` class provides an ``addCleanup`` method, which
1006
should be used instead of ``tearDown``. All the cleanups are run when the
1007
test finishes, regardless of whether it passes or fails. If one cleanup
1008
fails, later cleanups are still run.
1010
(The same facility is available outside of tests through
1011
``bzrlib.cleanup``.)
1017
Generally we prefer automated testing but sometimes a manual test is the
1018
right thing, especially for performance tests that want to measure elapsed
1019
time rather than effort.
1021
Simulating slow networks
1022
------------------------
1024
To get realistically slow network performance for manually measuring
1025
performance, we can simulate 500ms latency (thus 1000ms round trips)::
1027
$ sudo tc qdisc add dev lo root netem delay 500ms
1029
Normal system behaviour is restored with ::
1031
$ sudo tc qdisc del dev lo root
1033
A more precise version that only filters traffic to port 4155 is::
1035
tc qdisc add dev lo root handle 1: prio
1036
tc qdisc add dev lo parent 1:3 handle 30: netem delay 500ms
1037
tc qdisc add dev lo parent 30:1 handle 40: prio
1038
tc filter add dev lo protocol ip parent 1:0 prio 3 u32 match ip dport 4155 0xffff flowid 1:3 handle 800::800
1039
tc filter add dev lo protocol ip parent 1:0 prio 3 u32 match ip sport 4155 0xffff flowid 1:3 handle 800::801
1041
and to remove this::
1043
tc filter del dev lo protocol ip parent 1: pref 3 u32
1044
tc qdisc del dev lo root handle 1:
1046
You can use similar code to add additional delay to a real network
1047
interface, perhaps only when talking to a particular server or pointing at
1048
a VM. For more information see <http://lartc.org/>.
1051
872
.. |--| unicode:: U+2014
1054
vim: ft=rst tw=74 ai et sw=4
added
removed
doc/developers/testing.txt
