6
/* The driver function. This must return a CuSuite variable, which will
7
* then be used to actually run the tests. Essentially, all Suites are a
8
* collection of tests. The driver will take each Suite, and put it in a
9
* SuiteList, which is a collection of Suites.
11
CuSuite *testtime(void)
13
/* The actual suite, this must be created for each test program. Please
14
* give it a useful name, that will inform the user of the feature being
17
CuSuite *suite = CuSuiteNew("Test Time");
19
/* Each function must be added to the suite. Each function represents
20
* a single test. It is possible to test multiple features in a single
21
* function, although no tests currently do that.
23
SUITE_ADD_TEST(suite, test_now);
24
SUITE_ADD_TEST(suite, test_gmtstr);
25
SUITE_ADD_TEST(suite, test_localstr);
26
SUITE_ADD_TEST(suite, test_exp_get_gmt);
27
SUITE_ADD_TEST(suite, test_exp_get_lt);
28
SUITE_ADD_TEST(suite, test_imp_gmt);
29
SUITE_ADD_TEST(suite, test_rfcstr);
30
SUITE_ADD_TEST(suite, test_ctime);
31
SUITE_ADD_TEST(suite, test_strftime);
32
SUITE_ADD_TEST(suite, test_strftimesmall);
33
SUITE_ADD_TEST(suite, test_exp_tz);
34
SUITE_ADD_TEST(suite, test_strftimeoffset);
36
/* You must return the suite so that the driver knows which suites to
42
Building the full driver
43
------------------------
45
All you need to do to build the full driver is run:
56
Currently, some tests are known to fail in certain circumstances:
58
* 'testpoll' opens 64 sockets concurrently; ensure that resource
59
limits are high enough to allow this (using ulimit or limit); for
60
instance, Solaris <=2.7 and HP-UX 11.00 both set the limit to <=64 by
63
* 'testipsub' will tickle the Solaris 8 getaddrinfo() IPv6
64
bug, causing the test to hang. Configure with --disable-ipv6 if using
65
an unpatched Solaris 8 installation.
67
* The 'testdso' tests will not work if configured with
68
--disable-shared since the loadable modules cannot be built.
70
Running individual tests
71
---------------------------------
73
It is not possible to build individual tests, however it is possible to
74
run individual tests. When running the test suite, specify the name of the
75
tests that you want to run on the command line. For example:
77
./testall teststr testrand
79
Will run the Strings and Random generator tests.
81
Reading the test suite output
82
-----------------------------
84
Once you run the test suite, you will get output like:
88
Test Time: ............
90
16 tests run: 16 passed, 0 failed, 0 not implemented.
92
There are a couple of things to look at with this. First, if you look at the
93
first function in this document, you should notice that the string passed to
94
the CuSuiteNew function is in the output. That is why the string should
95
explain the feature you are testing.
97
Second, this test passed completely. This is obvious in two ways. First, and
98
most obvious, the summary line tells you that 16 tests were run and 16 tests
99
passed. However, the results can also be found in the lines above. Every
100
'.' in the output represents a passed test.
102
If a test fails, the output will look like:
106
Test Time: ..F.........
108
16 tests run: 15 passed, 1 failed, 0 not implemented.
110
This is not very useful, because you don't know which test failed. However,
111
once you know that a test failed, you can run the suite again, with the
112
-v option. If you do this, you will get something like:
116
Test Time: ..F.........
118
16 tests run: 15 passed, 1 failed, 0 not implemented.
120
1) test_localstr: assert failed
122
In this case, we know the test_localstr function failed, and there is an
123
Assert in this that failed (I modified the test to fail for this document).
124
Now, you can look at what that test does, and why it would have failed.
126
There is one other possible output for the test suite (run with -v):
130
Test Time: ..N.........
132
16 tests run: 15 passed, 0 failed, 1 not implemented.
134
Not Implemented tests:
136
Not Implemented tests:
137
1) test_localstr: apr_time_exp_lt not implemented on this platform
139
The 'N' means that a function has returned APR_ENOTIMPL. This should be
140
treated as an error, and the function should be implemented as soon as
143
Adding New test Suites to the full driver
144
-------------------------------------------
146
To add a new Suite to the full driver, you must make a couple of modifications.
148
1) Edit test_apr.h, and add the prototype for the function.
149
2) Edit testall.c, and add the function and name to the tests array.
150
3) Edit Makefile.in, and add the .lo file to the testall target.
152
Once those four things are done, your tests will automatically be added
158
There are a couple of rules for writing good tests for the test suite.
160
1) All tests can determine for themselves if it passed or not. This means
161
that there is no reason for the person running the test suite to interpret
162
the results of the tests.
163
2) Never use printf to add to the output of the test suite. The suite
164
library should be able to print all of the information required to debug
166
3) Functions should be tested with both positive and negative tests. This
167
means that you should test things that should both succeed and fail.
168
4) Just checking the return code does _NOT_ make a useful test. You must
169
check to determine that the test actually did what you expected it to do.
174
Finally, we will look at a quick test:
176
/* All tests are passed a CuTest variable. This is how the suite determines
177
* if the test succeeded or failed.
179
static void test_localstr(CuTest *tc)
185
rv = apr_time_exp_lt(&xt, now);
186
os_now = now / APR_USEC_PER_SEC;
188
/* If the function can return APR_ENOTIMPL, then you should check for it.
189
* This allows platform implementors to know if they have to implement
192
if (rv == APR_ENOTIMPL) {
193
CuNotImpl(tc, "apr_time_exp_lt");
196
/* It often helps to ensure that the return code was APR_SUCESS. If it
197
* wasn't, then we know the test failed.
199
CuAssertTrue(tc, rv == APR_SUCCESS);
201
/* Now that we know APR thinks it worked properly, we need to check the
202
* output to ensure that we got what we expected.
204
CuAssertStrEquals(tc, "2002-08-14 12:05:36.186711 -25200 [257 Sat] DST",
208
Notice, the same test can fail for any of a number of reasons. The first
209
test to fail ends the test.
214
CuTest is an open source test suite written by Asim Jalis. It has been
215
released under the zlib/libpng license. That license can be found in the
216
CuTest.c and CuTest.h files.
218
The version of CuTest that is included in the APR test suite has been modified
219
from the original distribution in the following ways:
221
1) The original distribution does not have a -v flag, the details are always
223
2) The NotImplemented result does not exist.
224
3) SuiteLists do not exist. In the original distribution, you can add suites
225
to suites, but it just adds the tests in the first suite to the list of tests
226
in the original suite. The output wasn't as detailed as I wanted, so I created
229
The first two modifications have been sent to the original author of CuTest,
230
but they have not been integrated into the base distribution. The SuiteList
231
changes will be sent to the original author soon.
233
The modified version of CuTest is not currently in any CVS or Subversion
234
server. In time, it will be hosted at rkbloom.net.
236
There are currently no docs for how to write tests, but the teststr and
237
testtime programs should give an idea of how it is done. In time, a document
238
should be written to define how tests are written.