203 validator custom message error causes test matcher error

[Mth0158]

@igorkasyanchuk here is a refacto proposition to make tests clearer.

In my opinion, switching to spec makes it really easier to read the tests, plus a big part of the community is using Rspec as their testing framework, so writing tests like specs makes even more sense.

I have just done the refacto for the size_validator_matcher as a POC, let me know if that's okay with you. Then, I was thinking of proceeding the following way:

• Refacto the whole gem tests => You validate
• Add new tests to mitigate regressions & fix Validator custom error message causes test matcher error #203 => You validate
  Then we could merge it

Let me know how I can be helpful

[igorkasyanchuk]

@Mth0158 I'm good with this refactoring, please continue 🚀

[Mth0158]

@igorkasyanchuk I think I'll:

• finish refactoring the 2 remaining matcher files (attached & content_type)
• add test for combined matchers (ie solving the Validator custom error message causes test matcher error #203 issue)

And maybe we can merge it if that's okay with you?
I will refactor the tests on the validators in another PR, since this PR is already adding a lot (as of now, it makes the test suite up to 360 from 232).

[igorkasyanchuk]

@Mth0158 I'm fine with it. If you want to merge - do it.

[Mth0158]

@igorkasyanchuk & @gr8bit I finished this PR.
I first added a lot of tests to make sure the big refacto in the last commit does not cause regressions.
The last commit is all about refactoring to simplify and DRY the code, it also fixes the issue #201 and add new inside features that can be reused somewhere else (the Validatable module, the Symobolizable module especially).
Let me know what you think of this, maybe it's better to only focus on the last commit since the other ones are all about cleaning and updating the test suite.

[Mth0158]

I still think we could go further in refactoring the code, but I'll stop here for this PR, it's big enough (make too big actually) 😅

[igorkasyanchuk]

@Mth0158 merged, thanks for the tremendous work. Do you think it worth to release a new version, or we can just wait.

[Mth0158]

@igorkasyanchuk thanks for the merge!
Maybe we can release it as a minor version so it definitely fixes the bug mentioned in #203 & #201 that some users had. I do not know exactly how you plan releases so let me know what you think is best in this use case.
It's not a critical release anyway, more something that will be useful to work with going on