In working on the IDI website, Cindy and I were having problems with different 
line heights, so I did a bit of research.

The UI Enhancer (used by UI Options to adjust the page) attaches a line-height 
CSS property to the body of the document. Currently, it uses a unit of 'em' for 
that line height. It seems that this is what was causing the problem.

When an absolute value is used for line-height (i.e. when a unit is specified, 
such as 'em' or 'px'), the computed line-height - based on the the font-size of 
the element where the line-height was specified - is inherited by all elements 
that don't have their own line-height set. In the case of UIO, this means that 
everything inherits the computed - and therefore fixed - line height based on 
the body font size. The result is that elements with larger fonts end up with 
too little line-spacing, requiring integrators to hard-code a line-height for 
those elements. Then, increasing the line-height with UIO doesn't affect those 
elements (this is what's happening in FLUID-4491). 

In contrast, when the line-height is a unitless number, it is treated as a 
scaling factor, and that factor is inherited. The factor will be used by each 
element to calculate the line height based on the element's own font size. I 
have posted a branch in which the 'em' is removed from the line-height:

        https://github.com/acheetham/infusion/tree/FLUID-4703-uie-line-height

This fixes the problems we were having in the IDI site, and allows us to remove 
the extra line-height properties in our UIO demo that were causing FLUID-4491. 
I've tested the fix across the sites that are currently using UIO. The change 
does result in some slight changes to line-spacing in some contexts, but 
nothing major.

I did have to adjust some of the tests to accommodate the change. Cindy was 
very helpful in figuring out what was going on and suggesting how to 
compensate: When unitless line-heights are used, a jQuery call to 
elem.css('lineHeight') returns the unitless factor in IE, but a calculated 
pixel value in all other browsers. The adjustment to the tests was to use the 
same check that UIEnhancer already uses for this issue.

-- 
Anastasia Cheetham     Inclusive Design Research Centre
[email protected]           Inclusive Design Institute
                                        OCAD University

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________________
fluid-work mailing list - [email protected]
To unsubscribe, change settings or access archives,
see http://lists.idrc.ocad.ca/mailman/listinfo/fluid-work

Reply via email to