[MathML] Increase visual space around fraction parts, italic variables, and operators
Created attachment 164937 [details] Patch
Comment on attachment 164937 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=164937&action=review Otherwise looks fine to me. I don't know if you want your MathML peeps to sign off or not. I'm not much help in evaluating the beauty of the Math. I'm ready to r+ with teh enum. > Source/WebCore/rendering/mathml/RenderMathMLFenced.h:44 > + RenderMathMLOperator* createMathMLOperator(UChar, bool isFence); Much better to use an enum here. Makes the callsites clearer. enum OperatorType { FenceOperator, NormalOperator? } createMathMLOperator can even have OperatorType = NormalOperator so most callsites don't have to pass anything if you like.
Created attachment 164949 [details] Patch
I added the enum. I cc'd David C. & Fred on the bug, but they're in Europe & may not see this soon. I don't think it needs their expert MathML review before landing. I'm just adding some space, making our layout look more similar to Firefox and TeX. I think we can't ask them to look at every tweak I make. We'll get plenty of reactions to layout issues from real users soon. :)
Comment on attachment 164949 [details] Patch LGTM. I agree, we don't need to consult the mathml council for every tweak. :)
Committed r129146: <http://trac.webkit.org/changeset/129146>
I think it would it would be nice in the future to move the tests to TestExpectations with a comment like "Needs rebaseline after rxxxx" for ports you don't rebaseline. That would avoid making our bots red until we do the rebaseline. Thanks in advance.
(In reply to comment #7) > I think it would it would be nice in the future to move the tests to TestExpectations with a comment like "Needs rebaseline after rxxxx" for ports you don't rebaseline. That would avoid making our bots red until we do the rebaseline. Thanks in advance. Good idea, I will do this.
From what I have checked, Firefox refers to an operator dictionary for Operator spacing. Are there any plans for Webkit to use something similar?
(In reply to comment #9) > From what I have checked, Firefox refers to an operator dictionary for Operator spacing. Are there any plans for Webkit to use something similar? Yes! See bug 99620 (or bug 99623).