Writing a new pipeline step and I keep guessing at the selection, running it, waiting, and finding out I was wrong. Twice now the wrong thing was "it selected nothing", which is not obvious from a green run.
Is there a way to ask what a command would do without doing it?
Accepted answer
--list. It prints the run targets and the tests that would run, then stops:
sdods run -p demo-shop --module cart -t @regression --listIt generates the specs to work out the list and then removes them along with the run directory, so it leaves nothing behind and you can run it as often as you like.
Two things it answers at once: which run targets your -l/-b selection produced, and how many tests are in them. Zero tests printed is the answer to your second problem.
And for the inverse question — "what tags exist that I could select" — sdods features list -p <slug> gives you module, scenario count, layers and tags per feature. I keep that output open while writing a tag expression.
Have a different question? Ask it.