ssshtest is designed to be practical and easy to use.
To use ssshtest in your project just source it in your test file
test -e ssshtest || wget -q https://raw.githubusercontent.com/ryanlayer/ssshtest/master/ssshtest
. ssshtest
Then write some tests:
run test_for_success python -c "print 'zzz: example success'"
assert_in_stdout "zzz"
run test_for_stderr python -c "sys.stderr.write('zzz: example failure')"
assert_in_stderr "xxx"
Then simply run the bash file that contains those lines.
$ bash mytests.sh
To run only certain tests, use:
$ bash mytests.sh test_for_success test_42
For an example of how to use travis continuous integration tests, our .travis.yml
run a block of code. This must precede any of the testing functions below.
- name for test.
- code to run
Assert that 2 things are equal. This does a string comparison so assert_equal "2" " 2" will fail.
- observed
- expected
assert_equal 42 $((21 + 21))
Assert that stdout is not empty
run test_stdout python -c "print 'zzz: example success'"
assert_stdout
Assert that stdout out contains this text.
run test_in_stdout python -c "print 'zzz: example success'"
assert_in_stdout "zzz"
- text to match
Assert that stdout is empty
run test_empty_stdout python -c "import sys; sys.stderr.write('aaa')"
assert_no_stdout
Assert that stderr is not empty
run test_stderr python -c "import sys; sys.stderr.write('zzz: example success')"
assert_stderr
Assert that stderr out contains this text.
- text to match
run test_in_stderr python -c "import sys; sys.stderr.write('zzz: example success')"
assert_in_stderr "zzz"
Assert that stderr is empty
run test_no_stderr python -c "print 'aaa'"
assert_no_stderr
Assert that the program exited with a particular code
- exit code
run test_exit_code python -c "import sys; sys.exit(33)"
assert_exit_code 33
Set STOP_ON_FAIL=1 after sourcing ssshtest to stop on the first error. Default is to continue running
$STDOUT_FILE is a file containing the $STDOUT from the last run command
$STDERR_FILE is a file containing the $STDERR from the last run command
ssshtest allows you to run a subset of the tests in your test file, and an
environment variable is set for to each test that is set to run. You can use
this variable to prevent lines associated with a test from running when the
test not set to run. For example, the following test creates the file
test.txt.
rm -f test.txt
run write_test \
python -c 'f=open("test.txt","w");f.write("test\n");f.close()'
assert_equal "test" $( cat test.txt )
If this test is not run the file is not created, and when the shell attempts to
resolve $( cat test.txt ) the error cat: test.txt: No such file or directory occurs.
You can prevent this error by putting a test-specific guard around the assert.
rm -f test.txt
run write_test \
python -c 'f=open("test.txt","w");f.write("test\n");f.close()'
if [ $write_test ]; then
assert_equal "test" $( cat test.txt )
fi
MIT LICENSE