Name conflicts in Haskell records

Question

Haskell doesn\'t have dot notation for record members. For each record member a compiler creates a function with the same name with a type RecType -> FieldType. This leads

Answer 1

Another way to avoid this problem is to use the lens package. It provides a makeFields template haskell function, which you can use like this:

{-# LANGUAGE FlexibleInstances      #-}
{-# LANGUAGE FunctionalDependencies #-}
{-# LANGUAGE MultiParamTypeClasses  #-}
{-# LANGUAGE TemplateHaskell        #-}
{-# LANGUAGE TypeSynonymInstances   #-}
import           Control.Lens

data A = A
  { _aText :: String
  }
makeFields ''A   -- Creates a lens x for each record accessor with the name _aX

data B = B
  { _bText  :: Int
  , _bValue :: Int
  }
-- Creates a lens x for each record accessor with the name _bX
makeFields ''B  

main = do
  let a = A "hello"
  let b = B 42 1

  -- (^.) is a function of lens which accesses a field (text) of some value (a)
  putStrLn $ "Text of a: " ++ a ^. text 
  putStrLn $ "Text of b: " ++ show (b ^. text)

If you don't want to use TemplateHaskell and lens, you can also do manually what lens automates using TemplateHaskell:

{-# LANGUAGE FlexibleInstances      #-}
{-# LANGUAGE FunctionalDependencies #-}
{-# LANGUAGE MultiParamTypeClasses  #-}
{-# LANGUAGE TypeSynonymInstances   #-}
data A = A
  { aText :: String
  }

data B = B
  { bText  :: Int
  , bValue :: Int
  }

-- A class for types a that have a "text" field of type t
class HasText a t | a -> t where

  -- An accessor for the text value
  text :: a -> t

-- Make our two types instances of those
instance HasText A String where text = aText
instance HasText B Int where text = bText

main = do
  let a = A "hello"
  let b = B 42 1
  putStrLn $ "Text of a: " ++ text a
  putStrLn $ "Text of b: " ++ show (text b)

But I can really recommend learning lens, as it also provides lots of other utilities, like modifying or setting a field.