2 subunit: A streaming protocol for test results
3 Copyright (C) 2005-2009 Robert Collins <robertc@robertcollins.net>
5 Licensed under either the Apache License, Version 2.0 or the BSD 3-clause
6 license at the users choice. A copy of both licenses are available in the
7 project source as Apache-2.0 and BSD. You may not use this file except in
8 compliance with one of these two licences.
10 Unless required by applicable law or agreed to in writing, software
11 distributed under these licenses is distributed on an "AS IS" BASIS, WITHOUT
12 WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
13 license you chose for the specific language governing permissions and
14 limitations under that license.
16 See the COPYING file for full details on the licensing of Subunit.
18 subunit reuses iso8601 by Michael Twomey, distributed under an MIT style
19 licence - see python/iso8601/LICENSE for details.
24 Subunit is a streaming protocol for test results. The protocol is human
25 readable and easily generated and parsed. By design all the components of
26 the protocol conceptually fit into the xUnit TestCase->TestResult interaction.
28 Subunit comes with command line filters to process a subunit stream and
29 language bindings for python, C, C++ and shell. Bindings are easy to write
32 A number of useful things can be done easily with subunit:
33 * Test aggregation: Tests run separately can be combined and then
34 reported/displayed together. For instance, tests from different languages
35 can be shown as a seamless whole.
36 * Test archiving: A test run may be recorded and replayed later.
37 * Test isolation: Tests that may crash or otherwise interact badly with each
38 other can be run seperately and then aggregated, rather than interfering
40 * Grid testing: subunit can act as the necessary serialisation and
41 deserialiation to get test runs on distributed machines to be reported in
44 Subunit supplies the following filters:
45 * tap2subunit - convert perl's TestAnythingProtocol to subunit.
46 * subunit2csv - convert a subunit stream to csv.
47 * subunit2pyunit - convert a subunit stream to pyunit test results.
48 * subunit2gtk - show a subunit stream in GTK.
49 * subunit2junitxml - convert a subunit stream to JUnit's XML format.
50 * subunit-diff - compare two subunit streams.
51 * subunit-filter - filter out tests from a subunit stream.
52 * subunit-ls - list info about tests present in a subunit stream.
53 * subunit-stats - generate a summary of a subunit stream.
54 * subunit-tags - add or remove tags from a stream.
56 Integration with other tools
57 ----------------------------
59 Subunit's language bindings act as integration with various test runners like
60 'check', 'cppunit', Python's 'unittest'. Beyond that a small amount of glue
61 (typically a few lines) will allow Subunit to be used in more sophisticated
67 Subunit has excellent Python support: most of the filters and tools are written
68 in python and there are facilities for using Subunit to increase test isolation
69 seamlessly within a test suite.
71 One simple way to run an existing python test suite and have it output subunit
72 is the module ``subunit.run``::
74 $ python -m subunit.run mypackage.tests.test_suite
76 For more information on the Python support Subunit offers , please see
77 ``pydoc subunit``, or the source in ``python/subunit/__init__.py``
82 Subunit has C bindings to emit the protocol, and comes with a patch for 'check'
83 which has been nominally accepted by the 'check' developers. See 'c/README' for
89 The C library is includable and usable directly from C++. A TestListener for
90 CPPUnit is included in the Subunit distribution. See 'c++/README' for details.
95 Similar to C, the shell bindings consist of simple functions to output protocol
96 elements, and a patch for adding subunit output to the 'ShUnit' shell test
97 runner. See 'shell/README' for details.
102 To ignore some failing tests whose root cause is already known::
104 subunit-filter --without 'AttributeError.*flavor'
110 Sample subunit wire contents
111 ----------------------------
115 success: test foo works.
117 failure: tar a file. [
120 foo.c:34 WARNING foo is not defined.
124 When run through subunit2pyunit::
128 ========================
133 foo.c:34 WARNING foo is not defined.
136 Subunit protocol description
137 ============================
139 This description is being ported to an EBNF style. Currently its only partly in
140 that style, but should be fairly clear all the same. When in doubt, refer the
141 source (and ideally help fix up the description!). Generally the protocol is
142 line orientated and consists of either directives and their parameters, or
143 when outside a DETAILS region unexpected lines which are not interpreted by
144 the parser - they should be forwarded unaltered.
146 test|testing|test:|testing: test LABEL
147 success|success:|successful|successful: test LABEL
148 success|success:|successful|successful: test LABEL DETAILS
150 failure: test LABEL DETAILS
152 error: test LABEL DETAILS
154 skip[:] test LABEL DETAILS
156 xfail[:] test LABEL DETAILS
157 uxsuccess[:] test LABEL
158 uxsuccess[:] test LABEL DETAILS
163 time: YYYY-MM-DD HH:MM:SSZ
166 DETAILS ::= BRACKETED | MULTIPART
167 BRACKETED ::= '[' CR UTF8-lines ']' CR
168 MULTIPART ::= '[ multipart' CR PART* ']' CR
169 PART ::= PART_TYPE CR NAME CR PART_BYTES CR
170 PART_TYPE ::= Content-Type: type/sub-type(;parameter=value,parameter=value)
171 PART_BYTES ::= (DIGITS CR LF BYTE{DIGITS})* '0' CR LF
173 unexpected output on stdout -> stdout.
174 exit w/0 or last test completing -> error
176 Tags given outside a test are applied to all following tests
177 Tags given after a test: line and before the result line for the same test
178 apply only to that test, and inherit the current global tags.
179 A '-' before a tag is used to remove tags - e.g. to prevent a global tag
180 applying to a single test, or to cancel a global tag.
182 The progress directive is used to provide progress information about a stream
183 so that stream consumer can provide completion estimates, progress bars and so
184 on. Stream generators that know how many tests will be present in the stream
185 should output "progress: COUNT". Stream filters that add tests should output
186 "progress: +COUNT", and those that remove tests should output
187 "progress: -COUNT". An absolute count should reset the progress indicators in
188 use - it indicates that two separate streams from different generators have
189 been trivially concatenated together, and there is no knowledge of how many
190 more complete streams are incoming. Smart concatenation could scan each stream
191 for their count and sum them, or alternatively translate absolute counts into
192 relative counts inline. It is recommended that outputters avoid absolute counts
193 unless necessary. The push and pop directives are used to provide local regions
194 for progress reporting. This fits with hierarchically operating test
195 environments - such as those that organise tests into suites - the top-most
196 runner can report on the number of suites, and each suite surround its output
197 with a (push, pop) pair. Interpreters should interpret a pop as also advancing
198 the progress of the restored level by one step. Encountering progress
199 directives between the start and end of a test pair indicates that a previous
200 test was interrupted and did not cleanly terminate: it should be implicitly
201 closed with an error (the same as when a stream ends with no closing test
202 directive for the most recently started test).
204 The time directive acts as a clock event - it sets the time for all future
205 events. The value should be a valid ISO8601 time.
207 The skip, xfail and uxsuccess outcomes are not supported by all testing
208 environments. In Python the testttools (https://launchpad.net/testtools)
209 library is used to translate these automatically if an older Python version
210 that does not support them is in use. See the testtools documentation for the
213 skip is used to indicate a test was discovered but not executed. xfail is used
214 to indicate a test that errored in some expected fashion (also know as "TODO"
215 tests in some frameworks). uxsuccess is used to indicate and unexpected success
216 where a test though to be failing actually passes. It is complementary to
225 * Update versions in configure.ac and python/subunit/__init__.py.
226 * Make PyPI and regular tarball releases. Upload the regular one to LP, the
228 * Push a tagged commit.