I have a set_up_before_script function which does a few things that might fail in certain circumstances (e.g. when a used command is not installed). When I set set -euo pipefail in these functions, bashunit simply exits with an error code, but without any message what went wrong. Is there a proposed workflow to handle errors in these functions?

For comparison: setting set -euo pipefail in a set_up function makes the test run continue on an error, with just the test marked as failed, but also without an error message.

Without setting set -euo pipefail, a failing command in a set_up or set_up_before_script function silently passes, without any notice of that failure.

IMO this is not ideal. In both cases bashunit should provide a useful error message when an error appears in a set_up or set_up_before_script function. And furthermore, the user should not be required to set set -euo pipefail themselves, or at least this should be noted as best practice somewhere in the docs (or any other proposed workflow).

Any opinions on this?