I think this is do-able without jQuery. The main issue is accounting for the variable height of the elements.
reference here: http://jsfiddle.net/uqZgt/1/
HTML:
<div class="container">
<div class="box-1">
This box has alot of content. This box has alot of content. This box has alot of content.
</div>
<div class="box-2">
This box has a little bit of content. This box has a little bit of content. This box has a little bit of content. This box has alot of content. This box has alot of content. This box has alot of content.
</div>
</div>
CSS:
.container {
width: 242px;
}
.container div {
width: 100px;
background: #ff0000;
float: left;
padding: 10px;
border-right: 2px solid #000
}
.box-1 + .box-2 {
border-right: none;
border-left: 2px solid #000
}
.box-1 ~ .box-2 {
margin-left: -2px
}
in this example, all divs in the .container
div have a 2px solid black border-right. However, an element with class box-2
which directly proceeds an element with .box-1
will have a 2px solid black border-left, and no border-right. So far this creates a 3px border in between the two elements.
Now, .box-1 ~ .box-2
selects any .box-1
that directly preceeds a .box-2
, and sets it's margin-left to -2px. This drags it's sibling two pixels to the left, which effectively makes the borders of both elements overlap.
The .container
div has a width equal to the sum of the width of the two elements (200px), plus padding (10px right and left = 20px) plus the width of one of the borders (2px). 242px, so the two elements fit perfectly.
No matter which div has more content, the border will appear to span the height of the div with the most content.