I stumbled across the Thrift framework not so long ago and have been intrigued ever since because of it’s promises. And it’s easy to fall in love with the marketing and promises of any framework when it tickles and uses the words and phrases you like to hear, not so?
So finally, I have the opportunity to test drive it: in a design that will become a product that will be released into production in the not so far future. So I want to make sure it’s ticking along as I would expect it to, especially if I want to test, implement, debug, deploy, maintain, scale and extend. And so far, i’m tres impressed.
On the server, I have the default Thrift server, Ruby flavoured, running on Mac OS X with a CSharp client on a Windows XP box (same LAN) and all with no extra tinkering; just as is, courtesy of the generator. Luverly stuff! And then, the implemented handler on the server uses ActiveRecord to handle the database lifting. No Rails: just Thrift and ActiveRecord. And it rocks (so far).
I’ll more than likely be documenting this implementation in more detail over time since the combination opens up many more opportunities in the space i love to
work play in. And that’s including the spike project source code.
References and gratitude to the following authors/posts which assisted in getting this going: