I have the following xpath statement in a selenium test:
//tbody/tr[td/span[text()=\'IPODate\']]/td[4]/input
It gets what I want but my tests a
the CSS locator is css=span.CN
for the span that has the text IPODate in it.
Saucelabs have a good explanaton of how it works on their blog
UPD:
Unfortunately CSS won't parse the entire tree when move up and down till it finds what it is after. This is the main reason why XPath is so slow. CSS finds an item and then can move laterally through the DOM. The CSS below will find the input box that is after the TD
that has class=TCN
which holds your span with text in the table.
CSS: table > tbody > tr > td.TCN + td + td > input