start moving code into standalone .hs files

Author: byorgey

CPiH.tex

@@ -206,21 +206,7 @@ \section{A basic solution pipeline}
 
 \begin{figure}
   \centering
-\begin{minted}{haskell}
-  import Control.Arrow ((>>>))
-
-  main = interact $ parse >>> map solve >>> format
-
-  parse :: String -> [[Integer]]
-  parse = _
-
-  solve :: [Integer] -> Integer
-  solve = _
-
-  format :: [Integer] -> String
-  format = _
-\end{minted}
-% $
+  \inputminted{haskell}{code/intro/Skeleton.hs}
   \caption{A skeleton solution for A Different Problem}
   \label{fig:skeleton-different}
 \end{figure}
@@ -299,10 +285,7 @@ \section{Parsing and formatting}
 
 \begin{figure}
   \centering
-\begin{minted}{haskell}
-  parse :: String -> [[Integer]]
-  parse = lines >>> map (words >>> map read)
-\end{minted}
+  \inputminted[firstline=5, lastline=6]{haskell}{code/intro/Different.hs}
   \caption{\h{parse} function for A Different Problem}
   \label{fig:parse-different}
 \end{figure}
@@ -347,6 +330,8 @@ \section{Parsing and formatting}
 need to use provided information about the length of each test case.
 We will discuss such parsing in \pref{chap:parsing}.
 
+\todo{include full solution for A Different Problem?}
+
 \section{Using partial functions}
 \label{sec:partial}
 
@@ -429,22 +414,13 @@ \section{Example: processing a list of strings}
 the list of words, updating an accumulator variable every time we see
 a string that does not contain the letter e.  An inexperienced
 Haskeller might produce something similar:
-\begin{minted}{haskell}
-lengthWithoutE :: [String] -> Int
-lengthWithoutE [] = 0
-lengthWithoutE (s:ss)
-  | 'e' `elem` s = lengthWithoutE ss
-  | otherwise = length s + lengthWithoutE ss
-\end{minted}
+\inputminted{haskell}{code/wholemeal/LengthWithoutE.hs}
 This code works, but there is a better way.  Instead of thinking in
 terms of processing the list one item at a time, an experienced
 Haskeller will think in terms of incrementally transforming the input
 into the desired output: first, filter out the strings we don't want;
 next, turn each string into its length; and finally sum the lengths.
-\begin{minted}{haskell}
-lengthWithoutE' :: [String] -> Int
-lengthWithoutE' = filter ('e' `notElem`) >>> map length >>> sum
-\end{minted}
+\inputminted{haskell}{code/wholemeal/LengthWithoutEWholemeal.hs}
 This is better in almost every way: it is shorter, easier to read and
 understand, easier to refactor, harder to get wrong, and easier to
 prove correct.
@@ -494,6 +470,9 @@ \subsection{Lists}
 lists with $10^5$ elements each is probably fine; having $10^5$ lists
 with two elements each is probably going to be too slow.
 
+\todo{pick two or three representative sample problems and discuss
+  their solutions?  One using scanl?}
+
 \kattis{foo, bar}
 
 \subsection{Sets}
@@ -546,26 +525,7 @@ \subsubsection{Example: Booking a Room}
 
 \begin{figure}
   \centering
-\begin{minted}{haskell}
-import           Control.Arrow ((>>>))
-import qualified Data.Set      as S
-
-main :: IO ()
-main = interact $ words >>> parse >>> solve
-
-data Hotel = Hotel
-  { numRooms :: Int
-  , booked   :: [Int]
-  }
-
-parse :: [String] -> Hotel
-parse (r:_:rs) = Hotel (read r) (map read rs)
-
-solve :: Hotel -> String
-solve (Hotel r rs) =
-  maybe "too late" show . S.lookupMin
-  $ S.fromList [1..r] `S.difference` S.fromList rs
-\end{minted}
+  \inputminted{haskell}{code/wholemeal/BookingARoom.hs}
   \caption{Booking a Room solution}
   \label{fig:bookingaroom}
 \end{figure}
@@ -1001,10 +961,12 @@ \section{ByteString}
 \end{minted}
 Now we're talking---this version completes in a blazing 0.04 seconds!
 
-% We can take these principles and use them to make a variant of the
-% `Scanner` module from last time which uses (lazy, ASCII) `ByteString`
-% instead of `String`, including the use of the `readInt` functions to
-% read `Int` values quickly.  You can [find it here](https://github.com/byorgey/comprog-hs/blob/master/ScannerBS.hs).
+We can take these principles and use them to make a variant of the
+\h{Scanner} module from last time which uses (lazy, ASCII)
+\h{ByteString} instead of \h{String}, including the use of the
+\h{readInt} functions to read \h{Int} values quickly.  You can [find
+it
+here](https://github.com/byorgey/comprog-hs/blob/master/ScannerBS.hs).
 
 \chapter{Monoids}
 \label{chap:monoids}

code/intro/Different.hs

@@ -0,0 +1,12 @@
+import Control.Arrow ((>>>))
+
+main = interact $ parse >>> map solve >>> format
+
+parse :: String -> [[Integer]]
+parse = lines >>> map (words >>> map read)
+
+solve :: [Integer] -> Integer
+solve [a,b] = abs (a - b)
+
+format :: [Integer] -> String
+format = map show >>> unlines

code/intro/Skeleton.hs

@@ -0,0 +1,12 @@
+import Control.Arrow ((>>>))
+
+main = interact $ parse >>> map solve >>> format
+
+parse :: String -> [[Integer]]
+parse = _
+
+solve :: [Integer] -> Integer
+solve = _
+
+format :: [Integer] -> String
+format = _

code/wholemeal/BookingARoom.hs

@@ -0,0 +1,18 @@
+import           Control.Arrow ((>>>))
+import qualified Data.Set      as S
+
+main :: IO ()
+main = interact $ words >>> parse >>> solve
+
+data Hotel = Hotel
+  { numRooms :: Int
+  , booked   :: [Int]
+  }
+
+parse :: [String] -> Hotel
+parse (r:_:rs) = Hotel (read r) (map read rs)
+
+solve :: Hotel -> String
+solve (Hotel r rs) =
+  maybe "too late" show . S.lookupMin
+  $ S.fromList [1..r] `S.difference` S.fromList rs

code/wholemeal/LengthWithoutE.hs

@@ -0,0 +1,5 @@
+lengthWithoutE :: [String] -> Int
+lengthWithoutE [] = 0
+lengthWithoutE (s:ss)
+  | 'e' `elem` s = lengthWithoutE ss
+  | otherwise = length s + lengthWithoutE ss

code/wholemeal/LengthWithoutEWholemeal.hs

@@ -0,0 +1,2 @@
+lengthWithoutE' :: [String] -> Int
+lengthWithoutE' = filter ('e' `notElem`) >>> map length >>> sum