5 \___|\___/|_| \_\_____|
10 perl (and a unix-style shell)
11 diff (when a test fails, a diff is shown)
12 stunnel (for HTTPS and FTPS tests)
13 OpenSSH or SunSSH (for SCP, SFTP and SOCKS4/5 tests)
15 TCP ports used by default:
17 - 8990 on localhost for HTTP tests
18 - 8991 on localhost for HTTPS tests
19 - 8994 on localhost for HTTP IPv6 tests
20 - 8992 on localhost for FTP tests
21 - 8995 on localhost for FTP (2) tests
22 - 8993 on localhost for FTPS tests
23 - 8996 on localhost for FTP IPv6 tests
24 - 8997 on localhost for TFTP tests
25 - 8999 on localhost for SCP/SFTP tests
26 - 9000 on localhost for SOCKS tests
28 The test suite runs simple FTP, HTTP and TFTP servers on these ports to
29 which it makes requests. For SSL tests, it runs stunnel to handle
30 encryption to the regular servers. For SSH, it runs a standard OpenSSH
31 server. For SOCKS4/5 tests SSH is used to perform the SOCKS functionality
32 and requires a SSH client and server.
34 The base port number shown above can be changed using runtests' -b option
35 to allow running more than one instance of the test suite simultaneously
39 'make test'. This builds the test suite support code and invokes the
40 'runtests.pl' perl script to run all the tests. Edit the top variables
41 of that script in case you have some specific needs, or run the script
42 manually (after the support code has been built).
44 The script breaks on the first test that doesn't do OK. Use -a to prevent
45 the script from abort on the first error. Run the script with -v for more
46 verbose output. Use -d to run the test servers with debug output enabled as
47 well. Specifying -k keeps all the log files generated by the test intact.
49 Use -s for shorter output, or pass test numbers to run specific tests only
50 (like "./runtests.pl 3 4" to test 3 and 4 only). It also supports test case
51 ranges with 'to', as in "./runtests 3 to 9" which runs the seven tests from
52 3 to 9. Any test numbers starting with ! are disabled, as are any test
53 numbers found in the file data/DISABLED (one per line).
55 Shell startup scripts:
56 Tests which use the ssh test server, SCP/SFTP/SOCKS tests, might be badly
57 influenced by the output of system wide or user specific shell startup scripts,
58 .bashrc, .profile, /etc/csh.cshrc, .login, /etc/bashrc, etc. which output text
59 messages or escape sequences on user login. When these shell startup messages
60 or escape sequences are output they might corrupt the expected stream of data
61 which flows to the sftp-server or from the ssh client which can result in bad
62 test behaviour or even prevent the test server from running.
64 If the test suite ssh or sftp server fails to start up and logs the message
65 'Received message too long' then you are certainly suffering the unwanted
66 output of a shell startup script. Locate, cleanup or adjust the shell script.
69 The test script will check that all allocated memory is freed properly IF
70 curl has been built with the CURLDEBUG define set. The script will
71 automatically detect if that is the case, and it will use the ../memanalyze
72 script to analyze the memory debugging output.
74 The -t option will enable torture testing mode, which runs each test
75 many times but causes a different memory allocation to fail on each
76 successive run. This tests the out of memory error handling code to
77 ensure that memory leaks do not occur even in those situations.
80 If a test case fails, you can conveniently get the script to invoke the
81 debugger (gdb) for you with the server running and the exact same command
82 line parameters that failed. Just invoke 'runtests.pl <test number> -g' and
83 then just type 'run' in the debugger to perform the command through the
86 If a test case causes a core dump, analyze it by running gdb like:
88 # gdb ../curl/src core
90 ... and get a stack trace with the gdb command:
95 All logs are generated in the logs/ subdirectory (it is emptied first
96 in the runtests.pl script). Use runtests.pl -k to keep the temporary files
100 All test cases are put in the data/ subdirectory. Each test is stored in the
101 file named according to the test number.
103 See FILEFORMAT for the description of the test case files.
106 gcc provides a tool that can determine the code coverage figures for
107 the test suite. To use it, configure curl with
108 CFLAGS='-fprofile-arcs -ftest-coverage -g -O0'. Make sure you run the normal
109 and torture tests to get more full coverage, i.e. do:
115 The graphical tool ggcov can be used to browse the source and create
116 coverage reports on *NIX hosts:
120 The text mode tool gcov may also be used, but it doesn't handle object files
121 in more than one directory very well.
124 The runtests.pl script provides some hooks to allow curl to be tested on a
125 machine where perl can not be run. The test framework in this case runs on
126 a workstation where perl is available, while curl itself is run on a remote
127 system using ssh or some other remote execution method. See the comments at
128 the beginning of runtests.pl for details.
132 So far, I've used this system:
139 500 - 599 libcurl source code tests, not using the curl command tool
141 700 - 799 SOCKS4 (even numbers) and SOCK5 (odd numbers)
142 1000 - 1999 miscellaneous*
143 2000 - x multiple sequential protocols per test case*
145 Since 30-apr-2003, there's nothing in the system that requires us to keep
146 within these number series, and those sections marked with * actually
147 contain tests for a variety of protocols. Each test case now specifies
148 its own server requirements, independent of test number.
152 * Add tests for TELNET, LDAP, DICT...
153 * SOCKS4/5 test deficiencies - no proxy authentication tests as SSH (the
154 test mechanism) doesn't support them