We are working towards making exercises stand-alone. That is to say: no more generating READMEs on the fly.
This will give maintainers more control over each individual exercise README, and it will also make some of the backend logic for delivering exercises simpler.
The README template uses the Go text/template package, and the default templates generate the same READMEs as we have been generating on the fly. See the documentation in [regenerating exercise readmes][regenerate-docs] for details.
The READMEs can be generated at any time using a new 'generate' command in configlet. This command has not yet landed in master or been released, but can be built from source in the generate-readmes branch on [configlet][].
[configlet]: https://github.com/exercism/configlet
[regenerate-docs]: https://github.com/exercism/docs/blob/master/maintaining-a-track/regenerating-exercise-readmes.md
I failed to notice with the initial commit that the first tests in these
files are just the "version sanity check". This should start learners
off with a useful first failure.
This allows learners to gradually approach exercise. However, unlike
commented tests, the Elm compiler is still able to infer type
information from the skipped tests.
* Update exercises to elm-test 2.0
* Update docs to mention `elm-test` again
* Update .travis.yml to the correct version of elm-test
* Conform to the `<| \() ->` convention
