MutationObserver: added test on special handling for inline styles

[just-boris]

Context: it was originally found as a bug in JSDOM: jsdom/jsdom#3305

But I think, the test case for that issue belongs to this repo

[annevk]

It's great to have tests for this, but I worry this is not well-defined. @zcorpan @emilio do you know if CSSOM covers attribute mutations to this level of detail?

[just-boris]

It is covered here: https://drafts.csswg.org/cssom/#dom-cssstyledeclaration-setproperty

If updated is true, update style attribute for the CSS declaration block.

When styles do not change, the attribute does not get set

[annevk]

Could you move this test into css/cssom?

[emilio]

@annevk Don't we have already css/cssom/cssstyledeclaration-mutationrecord-*.html and css/cssom/css-style-attr-decl-block.html for this? At least some of these are covered there, if not all.

[annevk]

Sorry, I should have grepped. Nice to see we gained test coverage for this.

[just-boris]

Checked the mentioned test suites. The cssText behavior is indeed covered by these lines:

wpt/css/cssom/css-style-attr-decl-block.html

Lines 73 to 74 in fee1162

	style.cssText = "z-index: 42;";
	assert_record_with_old_value("z-index: 42;", "changing cssText with the same content");

However, the primary part (no mutation records when style[something] gets set) was not covered before.

I can send a follow-up PR to restructure the tests if you will.