Blackbox testing of unix command line tools
cmdtest [-c=COMMAND] [--command=COMMAND] [--config=FILE] [--dump-config] [--dump-memory-profile=METHOD] [--dump-setting-names] [--generate-manpage=TEMPLATE] [-h] [--help] [--help-all] [-k] [--keep] [--list-config-files] [--log=FILE] [--log-keep=N] [--log-level=LEVEL] [--log-max=SIZE] [--log-mode=MODE] [--memory-dump-interval=SECONDS] [--no-default-configs] [--no-keep] [--no-timings] [--output=FILE] [-t=TEST] [--test=TEST] [--timings] [--version] [FILE]...
cmdtest black box tests Unix command line tools. Given some test scripts, their inputs, and expected outputs, it verifies that the command line produces the expected output. If not, it reports problems, and shows the differences.
Each test case foo consists of the following files:
foo.script
a script to run the test (this is required)
foo.stdin
the file fed to standard input
foo.stdout
the expected output to the standard output
foo.stderr
the expected output to the standard error
foo.exit
the expected exit code
foo.setup
a shell script to run before the test
foo.teardown
a shell script to run after test
Usually, a single test is not enough. All tests are put into the same directory, and they may share some setup and teardown code:
setup-once
a shell script to run once, before any tests
setup
a shell script to run before each test
teardown
a shell script to run after each test
teardown-once
a shell script to run once, after all tests
cmdtest is given the name of the directory with all the tests, or several such directories, and it does the following:
• execute setup-once
• for each test case (unique prefix foo):
– execute setup – execute foo.setup – execute the command, by running foo.script, and redirecting standard input to come from foo.stdin, and capturing standard output and error and exit codes – execute foo.teardown – execute teardown – report result of test: does exit code match foo.exit, standard output match foo.stdout, and standard error match foo.stderr?
• execute teardown-once
Except for foo.script, all of these files are optional. If a setup or teardown script is missing, it is simply not executed. If one of the standard input, output, or error files is missing, it is treated as if it were empty. If the exit code file is missing, it is treated as if it specified an exit code of zero.
The shell scripts may use the following environment variables:
DATADIR
a temporary directory where files may be created by the test
TESTNAME
name of the current test (will be empty for setup-once and teardown-once)
SRCDIR
directory from which cmdtest was launched
-c, --command=COMMAND
ignored for backwards compatibility
--generate-manpage=TEMPLATE
SUPPRESSHELP
-h, --help
show this help message and exit
-k, --keep
keep temporary data on failure
--no-keep
--no-timings
--output=FILE
write output to FILE, instead of standard output
-t, --test=TEST
run only TEST (can be given many times)
--timings
report how long each test takes
--version
show program's version number and exit
--config=FILE
add FILE to config files
--dump-config
write out the entire current configuration
--dump-setting-names
SUPPRESSHELP
--help-all
show all options
--list-config-files
SUPPRESSHELP
--no-default-configs
clear list of configuration files to read
--log=FILE
write log entries to FILE (default is to not write log files at all); use "syslog" to log to system log, or "none" to disable logging
--log-keep=N
keep last N logs (10)
--log-level=LEVEL
log at LEVEL, one of debug, info, warning, error, critical, fatal (default: debug)
--log-max=SIZE
rotate logs larger than SIZE, zero for never (default: 0)
--log-mode=MODE
set permissions of new log files to MODE (octal; default 0600)
--dump-memory-profile=METHOD
make memory profiling dumps using METHOD, which is one of: none, simple, meliae, or heapy (default: simple)
--memory-dump-interval=SECONDS
make memory profiling dumps at least SECONDS apart
To test that the echo(1) command outputs the expected string, create a file called echo-tests/hello.script containing the following content:
#!/bin/sh echo hello, world
Also create the file echo-tests/hello.stdout containing:
hello, world
Then you can run the tests:
$ cmdtest echo-tests test 1/1 1/1 tests OK, 0 failures
If you change the stdout file to be something else, cmdtest will report the differences:
$ cmdtest echo-tests FAIL: hello: stdout diff: --- echo-tests/hello.stdout 2011-09-11 19:14:47 +0100 +++ echo-tests/hello.stdout-actual 2011-09-11 19:14:49 +0100 @@ -1 +1 @@ -something else +hello, world test 1/1 0/1 tests OK, 1 failures
Furthermore, the echo-tests directory will contain the actual output files, and diffs from the expected files. If one of the actual output files is actually correct, you can actualy rename it to be the expected file. Actually, that's a very convenient way of creating the expected output files: you run the test, fixing things, until you've manually checked the actual output is correct, then you rename the file.