Hello!
Long time fan of Exercism, and I’m looking to get more involved.
I’d like to add the Simple Linked List exercise to the TS track.
Hello!
Long time fan of Exercism, and I’m looking to get more involved.
I’d like to add the Simple Linked List exercise to the TS track.
Thanks for the enthusiasm.
You can find a list of unimplemented exercises in the build status of the track.
Take a look at the Docs or the track specific CONTRIBUTING.md file for more on the specifics of adding an exercise. When you’re done and you open a PR it’ll be automatically closed by the bot so open a new forum post here, linking your PR, and one of the maintainers will take a look at it and reopen it.
Thank you for the reply!
Simple Linked List doesn’t seem to have a canonical-data.json
file, and I wasn’t sure if there was a reason for that, so I switched to another exercise that seemed to be a little more straightforward to implement (link to the forum post for that).
As far as Simple Linked List is concerned, I’d be happy to implement that once I have a better understanding of the whole workflow, I already have a working test suite, but I may have been missing something that’s needed to integrate it with the platform. Whatever is best for the language track and the Exercism platform.
Thanks again!
Most practice exercises should canonical-data.json files. Only about ten don’t. That isn’t to say those won’t have canonical test data in the future. In their absence, lots of tracks have already added Linked List and Simple Linked List. To add canonical data now would require those maintainers to identify a central set of tests and then also update their tracks’ tests to match. That’'d be a big effort at this point so I don’t think there’d be a lot of buy-in.
When I added Simple Linked List to the CoffeeScript track, I adapted the existing JavaScript implementation given the likely large overlap in folks enrolled in those tracks. TypeScript probably has a similar overlap so it’d make sense to adapt their Simple Linked List implementation as well if that helps.
I agree it would be a lot of effort … but so long as behavior is decently consistent across implementations, there might be decent buy in I can see maintainers being