ShoeQuery (better name needed) was born out of a desire to be able to find and manipulate elements in Shoes without having stashed away their locations ahead of time. It was inspired by jQuery's mechanisms for finding and manipulating DOM objects in Javascript.

The fundamental object (a ShoeQuery object) is pretty much an array of Shoes Elements, augmented with some extra functions. ShoeQuery objects are created via a helper (unsurprisingly named shoe_query) that lives in the Shoes::App object, and so should be accessible anywhere you want it. You can pass into that helper a selector, and it will return a ShoeQuery object containing the elements that match. You can also pass in an array of elements you already have handy, or an existing element, and it will also return a ShoeQuery object containing those elements. This form should be familiar for those who've used jQuery.

Right now, the things it lets you do are pretty limited, though a lot of the functionality jQuery specially provides already exists in Ruby. The one big feature is the ability to find elements by type, including css-like hierarchy. You can do this either by invoking shoe_query directly, or calling find on an existing shoe_query object (to search within its children). For example, given the demo app:

load 'shoe_query.rb' Shoes.app do app = Shoes.app do para "foo" f = flow do para "bar" stack do para 'blah' end end b = button "click me to change all paragraphs" b.click do shoe_query('para').each {|p| p.text = "all ps" } end b2 = button "click me to just change the paragraphs inside the flow" b2.click do shoe_query('flow').find('para').each {|p| p.text = "flow ps"} end b3 = button "click me to change the button inside the stack" b3.click do shoe_query('stack para').each {|p| p.text = "stack p"} end end

In this case, the first button modifies all three paragraph elements, while the latter two use two different mechanisms to scope their changes to within the flow and within the stack respectively