Falderal Test: Standard output and standard error

When we have a test that expects a successful result, the expected text is matched against standard output.

-> Functionality "Succeed" is implemented by shell command
-> "python fail.py '%(test-body-text)' bar 0"

-> Tests for functionality "Succeed"

| foo
= foo

If you wish to match the expected result against both standard output and standard error, it's up to you to redirect them.

-> Functionality "Succeed/All" is implemented by shell command
-> "python fail.py '%(test-body-text)' bar 0 2>&1"

-> Tests for functionality "Succeed/All"

| foo
= foo
= bar

When we have a test that expects an error result, the expected text is matched against standard error.

-> Functionality "Fail" is implemented by shell command
-> "python fail.py foo '%(test-body-text)' 1"

-> Tests for functionality "Fail"

| bar
? bar

If you wish to match the expected error against both standard output and standard error, it's up to you to redirect them.

-> Functionality "Fail/All" is implemented by shell command
-> "python fail.py foo '%(test-body-text)' 1 1>&2"

-> Tests for functionality "Fail/All"

| bar
? foo
? bar