[472] - Fix for Accept nullable parameter

[jahir-husain]

Fix for issue #472

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.03% 🎉

Comparison is base (76ea712) 90.08% compared to head (f0051ff) 90.11%.

❗ Current head f0051ff differs from pull request most recent head 3c7f264. Consider uploading reports for the commit 3c7f264 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #498      +/-   ##
==========================================
+ Coverage   90.08%   90.11%   +0.03%     
==========================================
  Files          76       76              
  Lines        1584     1589       +5     
==========================================
+ Hits         1427     1432       +5     
  Misses        157      157              

Files Changed	Coverage Δ
lib/json-schema/attribute.rb	100.00% <100.00%> (ø)
lib/json-schema/attributes/type.rb	100.00% <100.00%> (ø)
lib/json-schema/attributes/type_v4.rb	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[jahir-husain]

@bastelfreak - Can you review this, Please.

[bastelfreak]

@jahir-husain can you please rebase against our latest master branch and squash the commits down?

[belinskidima]

in case when type is boolean, you need to flatten the valid_classes

type = "boolean"
valid_classes = TYPE_CLASS_MAPPINGS.fetch(type) # => [TrueClass, FalseClass]
valid_classes = [valid_classes, NilClass] if nullable # =>  [[TrueClass, FalseClass], NilClass]
Array(valid_classes).any? { |c| data.is_a?(c) } # => #<TypeError: class or module required>

[jahir-husain]

Checking this @belinskidima

[bastelfreak]

@jahir-husain can you please take a look at the comment and rebase against our latest master (so we get rid of the merge commits).

[bastelfreak]

@jahir-husain please rebase instead of doing a merge.

[kntmrkm]

any update?

[bastelfreak]

@kntmrkm this PR needs a rebase. If you want you can checkout the code locally, rebase it against our Head and then submit a new PR.

[notEthan]

nullable has never been a JSON Schema keyword. It used to be a keyword of OpenAPI's schema, though isn't anymore. I don't know that a JSON Schema validator should be validating keywords that aren't part of JSON Schema.