haskell [[Char]]到[[Int]]

Question

I've got problem with a Haskell program. I'm trying to change [[Char]] to [[Int]] I've got

["2","2","1","2,2","1"] 

list of char list and I'm trying to change it to [[Int]]

[[2],[2],[1],[2,2],[1]]

I've tried

f :: [String] -> [Int]
f = map read

but it gives me

[2,2,1,*** Exception: Prelude.read: no parse

Can anybody help me with this?

Answer 1

The reason that this fails is because a string "2,2" can not be converted to an Int itself: this is a digit followed by a comma, followed by a digit. An Int is parsed by an optional minus sign, followed by some digits, and some extra possibilities like hexadecimal numbers, but let us ignore these for now.

The type signature you specify for f is however incorrect, based on the expected output. Your output type seems to be a list of lists of Int s, so [[Int]] . That means that you should specify f as:

f :: [String] -> 
f = ...

We thus need to read every String to an [Int] . We can not use read directly here, since read ing to an [Int] expects the string to start and end with square brackets . We can however add these manually like:

f :: [String] -> [[Int]]
f = map ()

or a point-free version:

f :: [String] -> [[Int]]
f = map ()

For example:

Prelude> f ["2","2","1","2,2","1"] 
[[2],[2],[1],[2,2],[1]]

Towards safer parsing with readMaybe

Parsing from String s like in the above way is of course not very " safe ", since it is possible that the String does not follow the format. We can make this more safe and use for example readMaybe :: Read a => String -> Maybe a :

import Text.Read()

f :: [String] -> [ [Int]]
f = map ( . ('[' :) . (++ "]"))

For example:

Prelude Text.Read> f ["2", "3;2", "4,7,3", "bla"]
[Just [2],Nothing,Just [4,7,3],Nothing]

we can omit the failed reads for example by using catMaybes :: [Maybe a] -> [a] :

import Data.Maybe()
import Text.Read(readMaybe)

f :: [String] -> [[Int]]
f =  . map (readMaybe . ('[' :) . (++ "]"))

For example:

Prelude Data.Maybe Text.Read> f ["2", "3;2", "4,7,3", "bla"]
[[2],[4,7,3]]

or as @dfeuer said, we can use traverse :: (Applicative f, Traversable t) => (a -> fb) -> ta -> f (tb) to return an [[Int]] result wrapped in a Just if all parsing succeeded, and Nothing otherwise:

import Text.Read(readMaybe)

f :: [String] ->  [[Int]]
f =  (readMaybe . ('[' :) . (++ "]"))

For example:

Prelude Text.Read> f ["2","2","1","2,2","1"] 
Just [[2],[2],[1],[2,2],[1]]
Prelude Text.Read> f ["2", "3;2", "4,7,3", "bla"]
Nothing

Parse with error messages with readEither

We can obtain an error message wrapped in a Left in case the parsing fails by using readEither :: Read a => String -> Either String a :

import Text.Read()

f :: [String] -> [ [Int]]
f = map ( . ('[' :) . (++ "]"))

For example:

Prelude Text.Read> f ["2", "3;2", "4,7,3", "bla"]
[Right [2],Left "Prelude.read: no parse",Right [4,7,3],Left "Prelude.read: no parse"]

and use traverse in the same way to obtain an error message wrapped in a Left or the complete result in a Right :

import Text.Read(readEither)

f :: [String] ->  [[Int]]
f =  (readEither . ('[' :) . (++ "]"))

For example:

Prelude Text.Read> f ["2","2","1","2,2","1"] 
Right [[2],[2],[1],[2,2],[1]]
Prelude Text.Read> f ["2", "3;2", "4,7,3", "bla"]
Left "Prelude.read: no parse"

Here, like @dfeuer says, it does not really shows much information. There are however parsers that can provide more informative parsing errors.