Adds a unit test to parse 3 digits versions

[ericcitaire]

See google/jsonnet#1014

[biojppm]

@ericcitaire hard to see from google/jsonnet#1014. Is it a duplicate of #292 (already fixed in master)?

[biojppm]

@ericcitaire I force-pushed a fix to your commit, and elaborated the tests.

I don't know what is the particular problem you are having (the linked bug does not provide any context and is unclear for me), but the tests now do a successful roundtrip of multi-dot version strings, both in YAML and JSON.

If the CI passes, this should be evidence that the problem is neither in ryml's parsing nor emitting. But it could be that you're using an older version where #292 is still active.

[codecov]

Codecov Report

Merging #302 (fc0cdf8) into master (23b05ec) will increase coverage by 0.02%.
The diff coverage is 97.49%.

@@            Coverage Diff             @@
##           master     #302      +/-   ##
==========================================
+ Coverage   97.20%   97.23%   +0.02%     
==========================================
  Files          78       78              
  Lines       18072    18358     +286     
==========================================
+ Hits        17567    17850     +283     
- Misses        505      508       +3     

Impacted Files	Coverage Δ
test/test_case.hpp	44.91% <ø> (ø)
test/test_group.cpp	97.78% <96.03%> (+0.35%)	⬆️
test/test_case.cpp	87.31% <100.00%> (+0.12%)	⬆️
test/test_group.hpp	100.00% <100.00%> (ø)
test/test_number.cpp	100.00% <100.00%> (ø)
test/test_parser.cpp	99.47% <100.00%> (+<0.01%)	⬆️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[biojppm]

Indeed the tests succeed (the 9 failures are unrelated). Merging now. Thanks!

[ericcitaire]

@ericcitaire hard to see from google/jsonnet#1014. Is it a duplicate of #292 (already fixed in master)?

Indeed, it seems to be a duplicate of #292. Sorry, I missed it.

So I guess google/jsonnet just have to rebuild with a recent version of rapidyaml.

Thanks !