There’s some infrastructure work that “the powers that be” need to perform. For example, here’s the moonscript PR for that. And then there’s some magic database work to make the track live for maintainers for testing.
I had @ErikSchierboom helping out since @iHiD is 100% on Jiki work these days.
I completed Leap and Darts. It was pretty slow going for me personally, but I didn’t notice any issues with the test runner yet.
I’ve just ran the script.
The language dialect being used is specified using directives like
{$mode ObjFPC}{$H+}
hello-world was inconsistent with all the later exercises. Fixing in PR 83
That’s a good start, but are we also updating the testing docs?
Updating testing docs: Document how to return values by keiravillekode · Pull Request #87 · exercism/free-pascal · GitHub
Looks good to me. Thanks
Three of the easiest exercises (darts pangram two-fer) now have approaches that showcase different language features.
https://exercism.org/docs/tracks/free-pascal/tests mentions that “All tests must have an “ok” status for the solution to be accepted.”
It’s unclear who/what is accepting the solution. Presumably, we’re telling the student here that the online editor runs all the tests every time.
It’s also unclear what “ok” status means. Perhaps we can provide a short example of what the test output might look like when run locally. Arturo does something like that at arturo/docs/TESTS.md at 20f48cb8b0871dbf4f116ffdf04292f76d615f03 · exercism/arturo · GitHub.
“So, remove this line before start to write your solution.” appears in https://exercism.org/docs/tracks/free-pascal/tests but that should be starting, not start.