在Haskell中从ByteString解析各种整数的通用有效方法

Question

Here's what I could come up with but I don't think it's very efficient or safe:

import qualified Data.ByteString.Char8 as B8

convert2Int = read . B8.unpack

Is there a better way to do this? I found functions in libraries that do it for the Int type, but not for other types like Int32 , Int64 , Word64 , etc.

Answer 1

The most straightforward way is probably decode from Data.Binary :

import qualified Data.ByteString.Lazy as BL
import Data.Binary (decode)

convertToIntegral :: (Binary a, Integral a) => BL.ByteString -> a
convertToIntegral = decode

There is also a decodeOrFail there, which produces an Either value if the decoding fails.

PS: How do I know that decode will work for the types you want? By reading the list of Binary instances .

Answer 2

An alternative way I found is to use the Attoparsec module :

import Data.Attoparsec.ByteString.Char8
import qualified Data.ByteString as B

convertToIntegral :: (Integral a) => B.ByteString -> Either String a
convertToIntegral = parseOnly decimal

Of course, this only works for strings without signs (unlike "-3" and "+5"). One may use "signed" instead of "decimal" in these cases.