Test how line-height works on <input type=text> and friends

[zcorpan]

See whatwg/html#5462

[dholbert]

I think the test expectations are wrong here, as discussed in https://bugzilla.mozilla.org/show_bug.cgi?id=1753537#c3 .

getComputedStyle returns the "resolved value", which is usually the computed value but is instead the used value in some cases, as discussed at https://drafts.csswg.org/cssom/#resolved-values . And line-height:[not normal] is one such case where it specifically returns the used value.

So: if the test is using getComputedStyle, then it should be expecting to get back the used value of line-height here; and given that the used value is specced as being clamped for these controls, then the test should be expecting that clamping to show up in the getComputedStyle returned value.

[zcorpan]

@dholbert thanks, you're correct. I'll fix the test.

[dholbert]

Looks good to me, just a few suggestions (the first of which is actually a suggestion for whatwg/html@c6440d4 to make it better-match these tests).

[zcorpan]

@dholbert thanks, clarified the spec in whatwg/html@40c9c5e

I also changed the test names to describe the expectations.

[dholbert]

@dholbert thanks, clarified the spec in whatwg/html@40c9c5e

FWIW I didn't mean to request a spec-clarification along those lines -- I don't think that spec-clarification (saying 'normal' is not supposed to be there in the used value) was strictly necessary, since line-height:normal is already explicitly defined as setting a numeric used value, here: https://www.w3.org/TR/CSS21/visudet.html#propdef-line-height :

normal
Tells user agents to set the used value to a "reasonable" value based on the font of the element. [...] We recommend a used value for 'normal' between 1.0 to 1.2. The computed value is 'normal'.

So normal is by-definition not something that should show up in the used value.

It's fine with the clarification, but if you end up doing further wordsmithing here later on and feel like removing that spec-clarification, I won't object. :)