Some exercises are missing templates in lib.rs

[tushartyagi]

I guess the issue is self explanatory. There are some exercises which are missing the template/function stub in the lib.rs file. For example, the lib.rs of etl is empty. #512 is about the same issue and it is still open.

I can raise PRs as and when I come across such exercises. But wanted to understand what would go inside the unimplemented!() description? Will the best description do, or are we following any pattern.

I've created a PR for this one: #550

[coriolinus]

The full list of exercises with empty stub files can be generated with

find exercises/ -name "lib.rs" -empty | cut -d'/' -f2

As of right now, that list is

• accumulate
• acronym
• allergies
• anagram
• binary-search
• bracket-push
• circular-buffer
• clock
• custom-set
• dominoes
• etl
• hexadecimal
• luhn-from
• luhn-trait
• minesweeper
• nucleotide-codons
• nucleotide-count
• parallel-letter-frequency
• phone-number
• pig-latin
• protein-translation
• queen-attack
• rectangles
• rna-transcription
• robot-name
• roman-numerals
• rotational-cipher
• sieve
• sublist
• tournament
• triangle
• wordy

In general, track maintainers will not add stubs for these exercises, but we're happy to review PRs which add appropriate stubs.

[nathanielknight]

I also noticed this. As discussed in #542, there are some exercises where adding even a boilerplate template with the appropriate types would take something away from the exercise. Even if the tests dictate what types you need to use, I think reading the tests and figuring out what you need is a useful exercise. In some of these cases, a little note explaining what you're being asked to do might be more appropriate.

That said, there are some exercises (custom-set is one I worked through recently) where you have to write out a whole bunch of boilerplate before you can get your first test to run. In that case, I might push a little harder for a template, especially if there are other exercises where you're doing the type-sleuthing yourself.