I am making a note of this for anyone that is not seeing any documentation for robots.txt and using patterns sometimes called globbing patterns. Here is a decent resource on using Robots and wildcards but to add to that which I do now you need something more powerful.
So you see this a lot e.g.
User-agent: *
Disallow: /search?s=*
But testing this within your google webmaster tools when you want to block
mysite.com/help/search?s=seo
OR
mysite.com/otherfolder/gobble/search/?s=tips
DOESNT WORK!
And you have to many folders...you need a globbing pattern to come to the rescue.
So use this pattern /**/ every time you need to represent a folder. Then you can simply block query strings more effectively with less code like this.
User-agent: *
Disallow: /**/**/?my_print
Disallow: /**/**/**/?my_print
I came back to this today as I needed a pattern to match some soft 404's that I simply don't want to be judged against. Google you judgy person..
So I needed to block (and similar)
site.com/resorts/tag/shopcountry/Cayman-Islands?order=rating-18
And this worked
Disallow: /**/**/**/*?order=*