xref: /freebsd/contrib/atf/doc/atf-test-case.4 (revision 2b8331622f0b212cf3bb4fc4914a501e5321d506)
1.\" Copyright (c) 2007 The NetBSD Foundation, Inc.
2.\" All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\"
13.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND
14.\" CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
15.\" INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
16.\" MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
17.\" IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS BE LIABLE FOR ANY
18.\" DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
20.\" GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
21.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
22.\" IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
23.\" OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN
24.\" IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25.Dd March 6, 2017
26.Dt ATF-TEST-CASE 4
27.Os
28.Sh NAME
29.Nm atf-test-case
30.Nd generic description of test cases
31.Sh DESCRIPTION
32A
33.Em test case
34is a piece of code that stress-tests a specific feature of the software.
35This feature is typically self-contained enough, either in the amount of
36code that implements it or in the general idea that describes it, to
37warrant its independent testing.
38Given this, test cases are very fine-grained, but they attempt to group
39similar smaller tests which are semantically related.
40.Pp
41A test case is defined by three components regardless of the language it is
42implemented in: a header, a body and a cleanup routine.
43The
44.Em header
45is, basically, a declarative piece of code that defines several
46properties to describe what the test case does and how it behaves.
47In other words: it defines the test case's
48.Em meta-data ,
49further described in the
50.Sx Meta-data
51section.
52The
53.Em body
54is the test case itself.
55It executes all actions needed to reproduce the test, and checks for
56failures.
57This body is only executed if the abstract conditions specified by the
58header are met.
59The
60.Em cleanup
61routine is a piece of code always executed after the body, regardless of
62the exit status of the test case.
63It can be used to undo side-effects of the test case.
64Note that almost all side-effects of a test case are automatically cleaned
65up by the library; this is explained in more detail in the rest of this
66document.
67.Pp
68It is extremely important to keep the separation between a test case's
69header and body well-defined, because the header is
70.Em always
71parsed, whereas the body is only executed when the conditions defined in
72the header are met and when the user specifies that test case.
73.Pp
74At last, test cases are always contained into test programs.
75The test programs act as a front-end to them, providing a consistent
76interface to the user and several APIs to ease their implementation.
77.Ss Results
78Upon termination, a test case reports a status and, optionally, a textual
79reason describing why the test reported such status.
80The caller must ensure that the test case really performed the task that its
81status describes, as the test program may be bogus and therefore providing a
82misleading result, e.g., providing a result that indicates success but the
83error code of the program says otherwise.
84.Pp
85The possible exit status of a test case are one of the following:
86.Bl -tag -width expectedXfailureXX
87.It expected_death
88The test case expects to terminate abruptly.
89.It expected_exit
90The test case expects to exit cleanly.
91.It expected_failure
92The test case expects to exit with a controller fatal/non-fatal failure.
93If this happens, the test program exits with a success error code.
94.It expected_signal
95The test case expects to receive a signal that makes it terminate.
96.It expected_timeout
97The test case expects to execute for longer than its timeout.
98.It passed
99The test case was executed successfully.
100The test program exits with a success error code.
101.It skipped
102The test case could not be executed because some preconditions were not
103met.
104This is not a failure because it can typically be resolved by adjusting
105the system to meet the necessary conditions.
106This is always accompanied by a
107.Em reason ,
108a message describing why the test was skipped.
109The test program exits with a success error code.
110.It failed
111An error appeared during the execution of the test case.
112This is always accompanied by a
113.Em reason ,
114a message describing why the test failed.
115The test program exits with a failure error code.
116.El
117.Pp
118The usefulness of the
119.Sq expected_*
120results comes when writing test cases that verify known failures caused,
121in general, due to programming errors (aka bugs).
122Whenever the faulty condition that the
123.Sq expected_*
124result is trying to cover is fixed, then the test case will be reported as
125.Sq failed
126and the developer will have to adjust it to match its new condition.
127.Pp
128It is important to note that all
129.Sq expected_*
130results are only provided as a
131.Em hint
132to the caller; the caller must verify that the test case did actually terminate
133as the expected condition says.
134.Ss Input/output
135Test cases are free to print whatever they want to their
136.Xr stdout 4
137and
138.Xr stderr 4
139file descriptors.
140They are, in fact, encouraged to print status information as they execute
141to keep the user informed of their actions.
142This is specially important for long test cases.
143.Pp
144Test cases will log their results to an auxiliary file, which is then
145collected by the test program they are contained in.
146The developer need not care about this as long as he uses the correct
147APIs to implement the test cases.
148.Pp
149The standard input of the test cases is unconditionally connected to
150.Sq /dev/zero .
151.Ss Meta-data
152The following metadata properties can be exposed via the test case's head:
153.Bl -tag -width requireXmachineXX
154.It descr
155Type: textual.
156Required.
157.Pp
158A brief textual description of the test case's purpose.
159Will be shown to the user in reports.
160Also good for documentation purposes.
161.It has.cleanup
162Type: boolean.
163Optional.
164.Pp
165If set to true, specifies that the test case has a cleanup routine that has
166to be executed by the runtime engine during the cleanup phase of the execution.
167This property is automatically set by the framework when defining a test case
168with a cleanup routine, so it should never be set by hand.
169.It ident
170Type: textual.
171Required.
172.Pp
173The test case's identifier.
174Must be unique inside the test program and should be short but descriptive.
175.It require.arch
176Type: textual.
177Optional.
178.Pp
179A whitespace separated list of architectures that the test case can be run
180under without causing errors due to an architecture mismatch.
181.It require.config
182Type: textual.
183Optional.
184.Pp
185A whitespace separated list of configuration variables that must be defined
186to execute the test case.
187If any of the required variables is not defined, the test case is
188.Em skipped .
189.It require.diskspace
190Type: integer.
191Optional.
192Specifies the minimum amount of available disk space needed by the test.
193The value can have a size suffix such as
194.Sq K ,
195.Sq M ,
196.Sq G
197or
198.Sq T
199to make the amount of bytes easier to type and read.
200.It require.files
201Type: textual.
202Optional.
203.Pp
204A whitespace separated list of files that must be present to execute the
205test case.
206The names of these files must be absolute paths.
207If any of the required files is not found, the test case is
208.Em skipped .
209.It require.machine
210Type: textual.
211Optional.
212.Pp
213A whitespace separated list of machine types that the test case can be run
214under without causing errors due to a machine type mismatch.
215.It require.memory
216Type: integer.
217Optional.
218Specifies the minimum amount of physical memory needed by the test.
219The value can have a size suffix such as
220.Sq K ,
221.Sq M ,
222.Sq G
223or
224.Sq T
225to make the amount of bytes easier to type and read.
226.It require.progs
227Type: textual.
228Optional.
229.Pp
230A whitespace separated list of programs that must be present to execute
231the test case.
232These can be given as plain names, in which case they are looked in the
233user's
234.Ev PATH ,
235or as absolute paths.
236If any of the required programs is not found, the test case is
237.Em skipped .
238.It require.user
239Type: textual.
240Optional.
241.Pp
242The required privileges to execute the test case.
243Can be one of
244.Sq root
245or
246.Sq unprivileged .
247.Pp
248If the test case is running as a regular user and this property is
249.Sq root ,
250the test case is
251.Em skipped .
252.Pp
253If the test case is running as root and this property is
254.Sq unprivileged ,
255the runtime engine will automatically drop the privileges if the
256.Sq unprivileged-user
257configuration property is set; otherwise the test case is
258.Em skipped .
259.It timeout
260Type: integral.
261Optional; defaults to
262.Sq 300 .
263.Pp
264Specifies the maximum amount of time the test case can run.
265This is particularly useful because some tests can stall either because they
266are incorrectly coded or because they trigger an anomalous behavior of the
267program.
268It is not acceptable for these tests to stall the whole execution of the
269test program.
270.Pp
271Can optionally be set to zero, in which case the test case has no run-time
272limit.
273This is discouraged.
274.It X- Ns Sq NAME
275Type: textual.
276Optional.
277.Pp
278A user-defined property named
279.Sq NAME .
280These properties are free form, have no special meaning within ATF, and can
281be specified at will by the test case.
282The runtime engine should propagate these properties from the test case to
283the end user so that the end user can rely on custom properties for test case
284tagging and classification.
285.El
286.Ss Environment
287Every time a test case is executed, several environment variables are
288cleared or reseted to sane values to ensure they do not make the test fail
289due to unexpected conditions.
290These variables are:
291.Bl -tag -width LCXMESSAGESXX
292.It Ev HOME
293Set to the work directory's path.
294.It Ev LANG
295Undefined.
296.It Ev LC_ALL
297Undefined.
298.It Ev LC_COLLATE
299Undefined.
300.It Ev LC_CTYPE
301Undefined.
302.It Ev LC_MESSAGES
303Undefined.
304.It Ev LC_MONETARY
305Undefined.
306.It Ev LC_NUMERIC
307Undefined.
308.It Ev LC_TIME
309Undefined.
310.It Ev TZ
311Hardcoded to
312.Sq UTC .
313.El
314.Ss Work directories
315The test program always creates a temporary directory
316and switches to it before running the test case's body.
317This way the test case is free to modify its current directory as it
318wishes, and the runtime engine will be able to clean it up later on in a
319safe way, removing any traces of its execution from the system.
320To do so, the runtime engine will perform a recursive removal of the work
321directory without crossing mount points; if a mount point is found, the
322file system will be unmounted (if possible).
323.Ss File creation mode mask (umask)
324Test cases are always executed with a file creation mode mask (umask) of
325.Sq 0022 .
326The test case's code is free to change this during execution.
327.Sh SEE ALSO
328.Xr atf-test-program 1
329