1 _ _ ____ _
2 ___| | | | _ \| |
3 / __| | | | |_) | |
4 | (__| |_| | _ <| |___
5 \___|\___/|_| \_\_____|
6
7The curl Test Suite
8
9 1. Running
10 1.1 Requires to run
11 1.2 Port numbers used by test servers
12 1.3 Test servers
13 1.4 Run
14 1.5 Shell startup scripts
15 1.6 Memory test
16 1.7 Debug
17 1.8 Logs
18 1.9 Test input files
19 1.10 Code coverage
20 1.11 Remote testing
21
22 2. Numbering
23 2.1 Test case numbering
24
25 3. Write tests
26 3.1 test data
27 3.2 curl tests
28 3.3 libcurl tests
29 3.4 unit tests
30
31 4. TODO
32 4.1 More protocols
33 4.2 SOCKS auth
34
35==============================================================================
36
371. Running
38
39 1.1 Requires to run
40
41 perl (and a unix-style shell)
42 python (and a unix-style shell)
43 diff (when a test fails, a diff is shown)
44 stunnel (for HTTPS and FTPS tests)
45 OpenSSH or SunSSH (for SCP, SFTP and SOCKS4/5 tests)
46 nghttpx (for HTTP/2 tests)
47
48 1.2 Port numbers used by test servers
49
50 - TCP/8990 for HTTP
51 - TCP/8991 for HTTPS
52 - TCP/8992 for FTP
53 - TCP/8993 for FTPS
54 - TCP/8994 for HTTP IPv6
55 - TCP/8995 for FTP (2)
56 - TCP/8996 for FTP IPv6
57 - UDP/8997 for TFTP
58 - UDP/8998 for TFTP IPv6
59 - TCP/8999 for SCP/SFTP
60 - TCP/9000 for SOCKS
61 - TCP/9001 for POP3
62 - TCP/9002 for IMAP
63 - TCP/9003 for SMTP
64 - TCP/9004 for SMTP IPv6
65 - TCP/9005 for RTSP
66 - TCP/9006 for RTSP IPv6
67 - TCP/9007 for GOPHER
68 - TCP/9008 for GOPHER IPv6
69 - TCP/9008 for HTTPS server with TLS-SRP support
70
71 1.3 Test servers
72
73 The test suite runs simple FTP, POP3, IMAP, SMTP, HTTP and TFTP stand-alone
74 servers on the ports listed above to which it makes requests. For SSL tests,
75 it runs stunnel to handle encryption to the regular servers. For SSH, it
76 runs a standard OpenSSH server. For SOCKS4/5 tests SSH is used to perform
77 the SOCKS functionality and requires a SSH client and server.
78
79 The base port number (8990), which all the individual port numbers are
80 indexed from, can be set explicitly using runtests.pl' -b option to allow
81 running more than one instance of the test suite simultaneously on one
82 machine, or just move the servers in case you have local services on any of
83 those ports.
84
85 The HTTP server supports listening on a Unix domain socket, the default
86 location is 'http.sock'.
87
88 1.4 Run
89
90 'make test'. This builds the test suite support code and invokes the
91 'runtests.pl' perl script to run all the tests. Edit the top variables
92 of that script in case you have some specific needs, or run the script
93 manually (after the support code has been built).
94
95 The script breaks on the first test that doesn't do OK. Use -a to prevent
96 the script from aborting on the first error. Run the script with -v for more
97 verbose output. Use -d to run the test servers with debug output enabled as
98 well. Specifying -k keeps all the log files generated by the test intact.
99
100 Use -s for shorter output, or pass test numbers to run specific tests only
101 (like "./runtests.pl 3 4" to test 3 and 4 only). It also supports test case
102 ranges with 'to', as in "./runtests 3 to 9" which runs the seven tests from
103 3 to 9. Any test numbers starting with ! are disabled, as are any test
104 numbers found in the files data/DISABLED or data/DISABLED.local (one per
105 line). The latter is meant for local temporary disables and will be ignored
106 by git.
107
108 When -s is not present, each successful test will display on one line the
109 test number and description and on the next line a set of flags, the test
110 result, current test sequence, total number of tests to be run and an
111 estimated amount of time to complete the test run. The flags consist of
112 these letters describing what is checked in this test:
113
114 s stdout
115 d data
116 u upload
117 p protocol
118 o output
119 e exit code
120 m memory
121 v valgrind
122
123 1.5 Shell startup scripts
124
125 Tests which use the ssh test server, SCP/SFTP/SOCKS tests, might be badly
126 influenced by the output of system wide or user specific shell startup
127 scripts, .bashrc, .profile, /etc/csh.cshrc, .login, /etc/bashrc, etc. which
128 output text messages or escape sequences on user login. When these shell
129 startup messages or escape sequences are output they might corrupt the
130 expected stream of data which flows to the sftp-server or from the ssh
131 client which can result in bad test behaviour or even prevent the test
132 server from running.
133
134 If the test suite ssh or sftp server fails to start up and logs the message
135 'Received message too long' then you are certainly suffering the unwanted
136 output of a shell startup script. Locate, cleanup or adjust the shell
137 script.
138
139 1.6 Memory test
140
141 The test script will check that all allocated memory is freed properly IF
142 curl has been built with the CURLDEBUG define set. The script will
143 automatically detect if that is the case, and it will use the
144 'memanalyze.pl' script to analyze the memory debugging output.
145
146 Also, if you run tests on a machine where valgrind is found, the script will
147 use valgrind to run the test with (unless you use -n) to further verify
148 correctness.
149
150 runtests.pl's -t option will enable torture testing mode, which runs each
151 test many times and makes each different memory allocation fail on each
152 successive run. This tests the out of memory error handling code to ensure
153 that memory leaks do not occur even in those situations. It can help to
154 compile curl with CPPFLAGS=-DMEMDEBUG_LOG_SYNC when using this option, to
155 ensure that the memory log file is properly written even if curl crashes.
156
157 1.7 Debug
158
159 If a test case fails, you can conveniently get the script to invoke the
160 debugger (gdb) for you with the server running and the exact same command
161 line parameters that failed. Just invoke 'runtests.pl <test number> -g' and
162 then just type 'run' in the debugger to perform the command through the
163 debugger.
164
165 1.8 Logs
166
167 All logs are generated in the log/ subdirectory (it is emptied first in the
168 runtests.pl script). Use runtests.pl -k to force it to keep the temporary
169 files after the test run since successful runs will clean it up otherwise.
170
171 1.9 Test input files
172
173 All test cases are put in the data/ subdirectory. Each test is stored in the
174 file named according to the test number.
175
176 See FILEFORMAT for the description of the test case files.
177
178 1.10 Code coverage
179
180 gcc provides a tool that can determine the code coverage figures for
181 the test suite. To use it, configure curl with
182 CFLAGS='-fprofile-arcs -ftest-coverage -g -O0'. Make sure you run the normal
183 and torture tests to get more full coverage, i.e. do:
184
185 make test
186 make test-torture
187
188 The graphical tool ggcov can be used to browse the source and create
189 coverage reports on *NIX hosts:
190
191 ggcov -r lib src
192
193 The text mode tool gcov may also be used, but it doesn't handle object files
194 in more than one directory very well.
195
196 1.11 Remote testing
197
198 The runtests.pl script provides some hooks to allow curl to be tested on a
199 machine where perl can not be run. The test framework in this case runs on
200 a workstation where perl is available, while curl itself is run on a remote
201 system using ssh or some other remote execution method. See the comments at
202 the beginning of runtests.pl for details.
203
2042. Numbering
205
206 2.1 Test case numbering
207
208 1 - 99 HTTP
209 100 - 199 FTP
210 200 - 299 FILE
211 300 - 399 HTTPS
212 400 - 499 FTPS
213 500 - 599 libcurl source code tests, not using the curl command tool
214 600 - 699 SCP/SFTP
215 700 - 799 SOCKS4 (even numbers) and SOCK5 (odd numbers)
216 800 - 849 IMAP
217 850 - 899 POP3
218 900 - 999 SMTP
219 1000 - 1299 miscellaneous
220 1300 - 1399 unit tests
221 1400 - 1499 miscellaneous
222 1500 - 1599 libcurl source code tests, not using the curl command tool
223 (same as 5xx)
224 1600 - 1699 unit tests
225 2000 - x multiple sequential protocols per test case
226
227 There's nothing in the system that *requires* us to keep within these number
228 series.
229
2303. Write tests
231
232 Here's a quick description on writing test cases. We basically have three
233 kinds of tests: the ones that test the curl tool, the ones that build small
234 applications and test libcurl directly and the unit tests that test
235 individual (possibly internal) functions.
236
237 3.1 test data
238
239 Each test has a master file that controls all the test data. What to read,
240 what the protocol exchange should look like, what exit code to expect and
241 what command line arguments to use etc.
242
243 These files are tests/data/test[num] where [num] is described in section 2
244 of this document, and the XML-like file format of them is described in the
245 separate tests/FILEFORMAT document.
246
247 3.2 curl tests
248
249 A test case that runs the curl tool and verifies that it gets the correct
250 data, it sends the correct data, it uses the correct protocol primitives
251 etc.
252
253 3.3 libcurl tests
254
255 The libcurl tests are identical to the curl ones, except that they use a
256 specific and dedicated custom-built program to run instead of "curl". This
257 tool is built from source code placed in tests/libtest and if you want to
258 make a new libcurl test that is where you add your code.
259
260 3.4 unit tests
261
262 Unit tests are tests in the 13xx sequence and they are placed in tests/unit.
263 There's a tests/unit/README describing the specific set of checks and macros
264 that may be used when writing tests that verify behaviors of specific
265 individual functions.
266
267 The unit tests depend on curl being built with debug enabled.
268
2694. TODO
270
271 4.1 More protocols
272
273 Add tests for TELNET, LDAP, DICT...
274
275 4.2 SOCKS auth
276
277 SOCKS4/5 test deficiencies - no proxy authentication tests as SSH (the
278 test mechanism) doesn't support them
279