rna-transcription 1.0.0: reject invalid inputs

[petertseng]

The invalid input cases originally came from
exercism/problem-specifications#119

Today's JSON file can be found at
https://github.com/exercism/x-common/blob/master/exercises/rna-transcription/canonical-data.json

This track was already using all the valid inputs and simply needed to
add these invalid inputs.

[petertseng]

Well, this is a different approach from, say, #238 which used unwrap. Note that https://github.com/exercism/xrust/blob/master/exercises/nucleotide-codons/tests/codons.rs does use assert_eq with Ok though. I think I like this better. The difference is what happens when I return an Err when I should return Ok. With assert_eq!(Ok(...), ...):

---- test_transcribes_cytosine_guanine stdout ----
	thread 'test_transcribes_cytosine_guanine' panicked at 'assertion failed: `(left == right)` (left: `Ok(RibonucleicAcid { nucleotides: "G" })`, right: `Err(())`)', tests/rna-transcription.rs:12

With assert_eq!(..., ....unwrap());

---- test_transcribes_cytosine_guanine stdout ----
	thread 'test_transcribes_cytosine_guanine' panicked at 'called `Result::unwrap()` on an `Err` value: ()', src/libcore/result.rs:859

So why don't I go ahead and say now that I like assert_eq(Ok(...), ...) better since it shows me both sides.

[ijanos]

I think using unwrap in an assert is an antipattern and we should avoid it.