Fix: Validate style attributes on the RHS

[sunesimonsen]

We currently ignore empty styles in RHS which is quite surprising. This PR
fixes it by validating the RHS styles.

[coveralls]

Pull Request Test Coverage Report for Build 832

• 6 of 6 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.08%) to 91.537%

---

Totals
Change from base Build 829:	0.08%
Covered Lines:	529
Relevant Lines:	559

---

💛 - Coveralls

[coveralls]

Pull Request Test Coverage Report for Build 832

• 6 of 6 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.08%) to 91.537%

---

Totals
Change from base Build 829:	0.08%
Covered Lines:	529
Relevant Lines:	559

---

💛 - Coveralls

[papandreou]

Looks good! It highlights that it would be good to introduce a proper CSS parser.

[sunesimonsen]

What are you implying there is something wrong with our regex machine 😜

[papandreou]

Not for a second :)

[alexjeffburke]

This looks really good. I think it's the right balance of not being overly prescriptive but still protecting the user at the point of checks.

Just one thing I wanted to double check, but in the case of an empty string style on the RHS I guess anything still satisfies? The only reason I ask is I believe that is the behaviour of "class" and keeping consistency would be a win.

[sunesimonsen]

Just one thing I wanted to double check, but in the case of an empty string style on the RHS I guess anything still satisfies? The only reason I ask is I believe that is the behaviour of "class" and keeping consistency would be a win.

That would pass as it would translate to the empty object on the RHS.

[sunesimonsen]

Released as unexpected-dom@4.8.0