Cells encapsulate display components. The brilliant part about them is that everything you need to actually render a bit of view is contained in one place, not crapped about your app.
If you feel like those things should also be packaged with the cell then cell assets is for you. All it does is let you get at other things which should be private data of the cell (like how it behaves)
Let's say I've got some cell:
class MyCell < Cell::Rails def hello render end end
that has a directory structure like:
my/hello.html.erb /init.js /dragndrop.js /mycell.css /icon.png
MyCell.assets("*.js") #=> ['dragndrop.js','init.js']
if you pass it a block it will give you the content of each asset. For example, let's just read the init code:
MyCell.assets("init.js") do |content| # content contains in init.js end
matching assets are always returned in alphabetically sorted order so that you don't run into ordering problems when using different operating systems.
We use this as part of system that loads cells in isolation and so it is important that each cell be aware of all the bits that it needs to run. Perhaps this will be useful to other folks in the future. What you do with cell assets is your business, but the point is that the cell's assets are it's business and ain't nobody else's