Im perplexed by this one.
jQuery.height() is coming back with different values in Firefox and Chrome. Measuring the pixels on-screen indicates that of the two, Chrome ap
Firefox and in general, Gecko based browsers differ from others in that they try to do subpixel layout and rendering.
This can make life hard especially if you work with table cells that are sized based on their contents.
IE, Webkit and Gecko can report different dimensions - the latter reporting some exotic fractional sizes as well.
As for the bug: after struggling with something similar - measuring dimensions of dynamically sized table cells - for a while, I had ended up special casing for fractional numbers.
As it influences positions as well as dimensions, some people ended up using clientWidth instead of jQuery's 'innerWidth()', while some just use parseInt() on the returned results.
(Search for fract
in the sources.)
If that is not satisfactory, you may try the following:
It might still occur that the borders are misaligned - this can be the case if the position of the primary and secondary tables are not the same (e.g.: primary top: 123.75px, secondary top: 123px.) In that case, placing the tables in a common container or just repositioning to integer coordinates may help.
Good luck...
Edit: As far as I remember, because of some IE and border-collapse anomalies, I resorted to the (deprecated) cellspacing property - that should be placed on the table element into HTML, not CSS - and specified separate borders with a dimension of 0
in CSS. I placed the whole table in a div with a nice background-color.
The whole thing had the effect as if the table borders were 1px, collapsed having the color of the background div....