Got the itch to start studying Haskell again. Best to scratch it. I’m a bit sick of gluing together bits of Javascript code that don’t look like they should be together in the same script, then wondering why in the world it works. Granted, I did enjoy coding everything in the cloud, and my ultimate hope is for the same treatment be given to Haskell.

I keep getting drawn to Haskell’s simplicity, consistency, structural elegance, and perfect typing system, even though it still causes me a great deal of mental strain to understand it. And I found that after being away from it for several months, even the simplest concepts have to be re-learned.

So I did two things. One, I began re-reading Haskell guides like “Real World Haskell” and “Learn you a Haskell for Great Good!“. There are tons of very helpful resources online.

The second thing I did is to give “Programming Praxis” a try. In the past I used Project Euler, but having solved 108+ problems, I’m getting tired of the same old, same old. I’m sure I’ll get to the same point with Praxis, but for now its giving me a change.

**Praxis 1 – Create a RPN calculator**:

Loved my HP 48sx+ calculator in Engineering. Sad day when it kak’d. Still have trouble using regular algebraic calculators. So this one was is right up my alley. Lots of sample code online, but I started noticing that most code used function composition, rather than using variables. Up until now I always avoided function composition, but it really does make things clearer.

```
import Data.List
solveRPN :: String -> Float
solveRPN = head . foldl calcme [] . words
where
calcme (x:y:ys) "*" = (x * y):ys
calcme (x:y:ys) "+" = (x + y):ys
calcme (x:y:ys) "-" = (y - x):ys
calcme (x:y:ys) "/" = (y / x):ys
calcme xs numStr = read numStr:xs
main = do
putStrLn "Enter calculation: "
input <- getLine
putStrLn $ "Result = " ++ (show $ solveRPN input)
```