1 Writing TAP Tests
2
3Introduction
4
5 This is a guide for users of the C TAP Harness package or similar
6 TAP-based test harnesses explaining how to write tests. If your
7 package uses C TAP Harness as the test suite driver, you may want to
8 copy this document to an appropriate file name in your test suite as
9 documentation for contributors.
10
11About TAP
12
13 TAP is the Test Anything Protocol, a protocol for communication
14 between test cases and a test harness. This is the protocol used by
15 Perl for its internal test suite and for nearly all Perl modules,
16 since it's the format used by the build tools for Perl modules to run
17 tests and report their results.
18
19 A TAP-based test suite works with a somewhat different set of
20 assumptions than an xUnit test suite. In TAP, each test case is a
21 separate program. That program, when run, must produce output in the
22 following format:
23
24 1..4
25 ok 1 - the first test
26 ok 2
27 # a diagnostic, ignored by the harness
28 not ok 3 - a failing test
29 ok 4 # skip a skipped test
30
31 The output should all go to standard output. The first line specifies
32 the number of tests to be run, and then each test produces output that
33 looks like either "ok <n>" or "not ok <n>" depending on whether the
34 test succeeded or failed. Additional information about the test can
35 be provided after the "ok <n>" or "not ok <n>", but is optional.
36 Additional diagnostics and information can be provided in lines
37 beginning with a "#".
38
39 Processing directives are supported after the "ok <n>" or "not ok <n>"
40 and start with a "#". The main one of interest is "# skip" which says
41 that the test was skipped rather than successful and optionally gives
42 the reason. Also supported is "# todo", which normally annotates a
43 failing test and indicates that test is expected to fail, optionally
44 providing a reason for why.
45
46 There are three more special cases. First, the initial line stating
47 the number of tests to run, called the plan, may appear at the end of
48 the output instead of the beginning. This can be useful if the number
49 of tests to run is not known in advance. Second, a plan in the form:
50
51 1..0 # skip entire test case skipped
52
53 can be given instead, which indicates that this entire test case has
54 been skipped (generally because it depends on facilities or optional
55 configuration which is not present). Finally, if the test case
56 encounters a fatal error, it should print the text:
57
58 Bail out!
59
60 on standard output, optionally followed by an error message, and then
61 exit. This tells the harness that the test aborted unexpectedly.
62
63 The exit status of a successful test case should always be 0. The
64 harness will report the test as "dubious" if all the tests appeared to
65 succeed but it exited with a non-zero status.
66
67Writing TAP Tests
68
69 Environment
70
71 One of the special features of C TAP Harness is the environment that
72 it sets up for your test cases. If your test program is called under
73 the runtests driver, the environment variables SOURCE and BUILD will
74 be set to the top of the test directory in the source tree and the top
75 of the build tree, respectively. You can use those environment
76 variables to locate additional test data, programs and libraries built
77 as part of your software build, and other supporting information
78 needed by tests.
79
80 The C and shell TAP libraries support a test_file_path() function,
81 which looks for a file under the build tree and then under the source
82 tree, using the BUILD and SOURCE environment variables, and return the
83 full path to the file. This can be used to locate supporting data
84 files.
85
86 Perl
87
88 Since TAP is the native test framework for Perl, writing TAP tests in
89 Perl is very easy and extremely well-supported. If you've never
90 written tests in Perl before, start by reading the documentation for
91 Test::Tutorial and Test::Simple, which walks you through the basics,
92 including the TAP output syntax. Then, the best Perl module to use
93 for serious testing is Test::More, which provides a lot of additional
94 functions over Test::Simple including support for skipping tests,
95 bailing out, and not planning tests in advance. See the documentation
96 of Test::More for all the details and lots of examples.
97
98 C TAP Harness can run Perl test scripts directly and interpret the
99 results correctly, and similarly the Perl Test::Harness module and
100 prove command can run TAP tests written in other languages using, for
101 example, the TAP library that comes with C TAP Harness. You can, if
102 you wish, use the library that comes with C TAP Harness but use prove
103 instead of runtests for running the test suite.
104
105 C
106
107 C TAP Harness provides a basic TAP library that takes away most of the
108 pain of writing TAP test cases in C. A C test case should start with
109 a call to plan(), passing in the number of tests to run. Then, each
110 test should use is_int(), is_string(), is_double(), or is_hex() as
111 appropriate to compare expected and seen values, or ok() to do a
112 simpler boolean test. The is_*() functions take expected and seen
113 values and then a printf-style format string explaining the test
114 (which may be NULL). ok() takes a boolean and then the printf-style
115 string.
116
117 Here's a complete example test program that uses the C TAP library:
118
119 #include <stddef.h>
120 #include <tap/basic.h>
121
122 int
123 main(void)
124 {
125 plan(4);
126
127 ok(1, "the first test");
128 is_int(42, 42, NULL);
129 diag("a diagnostic, ignored by the harness");
130 ok(0, "a failing test");
131 skip("a skipped test");
132
133 return 0;
134 }
135
136 This test program produces the output shown above in the section on
137 TAP and demonstrates most of the functions. The other functions of
138 interest are sysdiag() (like diag() but adds strerror() results),
139 bail() and sysbail() for fatal errors, skip_block() to skip a whole
140 block of tests, and skip_all() which is called instead of plan() to
141 skip an entire test case.
142
143 The C TAP library also provides plan_lazy(), which can be called
144 instead of plan(). If plan_lazy() is called, the library will keep
145 track of how many test results are reported and will print out the
146 plan at the end of execution of the program. This should normally be
147 avoided since the test may appear to be successful even if it exits
148 prematurely, but it can make writing tests easier in some
149 circumstances.
150
151 Complete API documentation for the basic C TAP library that comes with
152 C TAP Harness is available at:
153
154 <http://www.eyrie.org/~eagle/software/c-tap-harness/>
155
156 It's common to need additional test functions and utility functions
157 for your C tests, particularly if you have to set up and tear down a
158 test environment for your test programs, and it's useful to have them
159 all in the libtap library so that you only have to link your test
160 programs with one library. Rather than editing tap/basic.c and
161 tap/basic.h to add those additional functions, add additional *.c and
162 *.h files into the tap directory with the function implementations and
163 prototypes, and then add those additional objects to the library.
164 That way, you can update tap/basic.c and tap/basic.h from subsequent
165 releases of C TAP Harness without having to merge changes with your
166 own code.
167
168 Libraries of additional useful TAP test functions are available in
169 rra-c-util at:
170
171 <http://www.eyrie.org/~eagle/software/rra-c-util/>
172
173 Some of the code there is particularly useful when testing programs
174 that require Kerberos keys.
175
176 If you implement new test functions that compare an expected and seen
177 value, it's best to name them is_<something> and take the expected
178 value, the seen value, and then a printf-style format string and
179 possible arguments to match the calling convention of the functions
180 provided by C TAP Harness.
181
182 Shell
183
184 C TAP Harness provides a library of shell functions to make it easier
185 to write TAP tests in shell. That library includes much of the same
186 functionality as the C TAP library, but takes its parameters in a
187 somewhat different order to make better use of shell features.
188
189 The libtap.sh file should be installed in a directory named tap in
190 your test suite area. It can then be loaded by tests written in shell
191 using the environment set up by runtests with:
192
193 . "$SOURCE"/tap/libtap.sh
194
195 Here is a complete test case written in shell which produces the same
196 output as the TAP sample above:
197
198 #!/bin/sh
199
200 . "$SOURCE"/tap/libtap.sh
201 cd "$BUILD"
202
203 plan 4
204 ok 'the first test' true
205 ok '' [ 42 -eq 42 ]
206 diag a diagnostic, ignored by the harness
207 ok '' false
208 skip 'a skipped test'
209
210 The shell framework doesn't provide the is_* functions, so you'll use
211 the ok function more. It takes a string describing the text and then
212 treats all of its remaining arguments as a condition, evaluated the
213 same way as the arguments to the "if" statement. If that condition
214 evaluates to true, the test passes; otherwise, the test fails.
215
216 The plan, plan_lazy, diag, and bail functions work the same as with
217 the C library. skip takes a string and skips the next test with that
218 explanation. skip_block takes a count and a string and skips that
219 many tests with that explanation. skip_all takes an optional reason
220 and skips the entire test case.
221
222 Since it's common for shell programs to want to test the output of
223 commands, there's an additional function ok_program provided by the
224 shell test library. It takes the test description string, the
225 expected exit status, the expected program output, and then treats the
226 rest of its arguments as the program to run. That program is run with
227 standard error and standard output combined, and then its exit status
228 and output are tested against the provided values.
229
230 A utility function, strip_colon_error, is provided that runs the
231 command given as its arguments and strips text following a colon and a
232 space from the output (unless there is no whitespace on the line
233 before the colon and the space, normally indicating a prefix of the
234 program name). This function can be used to wrap commands that are
235 expected to fail with output that has a system- or locale-specific
236 error message appended, such as the output of strerror().
237
238License
239
240 This file is part of the documentation of C TAP Harness, which can be
241 found at <http://www.eyrie.org/~eagle/software/c-tap-harness/>.
242
243 Copyright 2010 Russ Allbery <eagle@eyrie.org>
244
245 Copying and distribution of this file, with or without modification,
246 are permitted in any medium without royalty provided the copyright
247 notice and this notice are preserved. This file is offered as-is,
248 without any warranty.
249