>Have a click - you'l see the result of choice use of XSLT, REXML, YAML, >XHTML, XML, and fractals.
So I click on, say tropism (I click on the paper&quil icon) and then some code appears and below that I see what looks like a screenshot of output from FleaGL. So is that a screenshot generated when I click the paper&quil icon or is it just a stored image?
This all looks very cool, just trying to figure out if it's cool or WAY COOL.
> So I click on, say tropism (I click on the paper&quil icon) and then some > code appears and below that I see what looks like a screenshot of output > from FleaGL. So is that a screenshot generated when I click the > paper&quil icon or is it just a stored image?
It's an image stored from the last time the test ran. If the file were not read-only, and if you changed the source and hit Test, the back-end would render on the server, and you'd get a new fractal shape.
The problem with testing is it uses the same tools as cracking, so I can't let you change the Ruby source; you might do something naughty. I will work on that.
High Moon Studios uses this rig to test DarkWatch. You change Lua script, hit Test, and their game runs on the server and collects a screenshot of the hero springing some trap in the game.
> This all looks very cool, just trying to figure out if it's cool or WAY > COOL.
Forget games. Imagine if you need to localize a GUI to 10 languages. You configure the same test suite to run in each language, taking screen shots as it goes. Then you tell your linguists to review all the outputs, looking for all the little typos and missing ` marks that we gringos wouldn't notice.
If you must review appearances manually, at least time- and space-shift them away from the bench version of the application!
|
