haskell - is this a bug of 32bit GHC on MAC -

to make long stroy short, code parse json file using aeson

here 2 pieces of code:

a.hs

{-# language overloadedstrings #-}  import data.aeson import qualified data.bytestring.lazy.char8 c import control.monad import control.applicative  data auctioninfo = auctioninfo {     realm :: realm ,     alliance :: auctions  ,     horde ::  auctions  ,     neutral :: auctions   }   deriving (show )  instance fromjson auctioninfo    parsejson (object o) =       r <- o .: "realm" >>= parsejson       <- o .: "alliance" >>= parsejson       h <- o .: "horde" >>= parsejson       n <- o .: "neutral" >>= parsejson       return $  auctioninfo r h n    parsejson _ = mzero  data realm = realm { name2 :: string , slug:: string} deriving (show ) instance fromjson realm       parsejson (object o) = realm <$>             o .: "name" <*>             o .: "slug"       parsejson _ = mzero  data auctions = auctions {auctions :: [auc]} deriving (show) instance fromjson auctions         parsejson (object o ) = auctions <$> o.: "auctions"       parsejson _ = mzero  data auc = auc {     auc :: integer,     itme :: int,     owner ::  string,     bid :: integer,     buyout ::integer,     quantity :: int,     timeleft :: string,     rand :: integer,     seed :: integer     } deriving (show )  instance fromjson auc       parsejson (object o ) = auc <$>             o .: "auc" <*>             o .: "item" <*>             o .: "owner" <*>             o .: "bid" <*>             o .: "buyout" <*>             o .: "quantity" <*>             o .: "timeleft" <*>             o .: "rand" <*>             o .: "seed"       parsejson _ = mzero  main =     au<- c.readfile "a.json"     let x = decode au :: maybe auctioninfo     case x of         ->             {-putstrln.show $ a-}             putstrln .show.length.auctions.alliance $             putstrln "ok"         nothing -> putstrln "fail" 

my json test file

and test steps:

1. save code , , name a.hs (or want)
2. save test data ,name a.json (do not change name)
3. if have not install aeson, $ cabal install aseon
4. $ ghc a.hs -o a
5. $ ./a

what output "fail".

and when run command $ runghc a.hs few times , got ok , fail mixed together.

i have tried code on linux , 64bit mac ghc, output ok expected.

one of friends has tried code on 32bit mac ghc, fail too. , told me played black magic code , changed 1 line into

 let x = decode $(c.pack. c.unpack) au :: maybe auctioninfo 

then output ok. when did same black magic, output still fail.

i want make sure bug or bug of ghc, or how can determine that.

i'm not sure if behaviour related this, absolutely shouldn't use data.bytestring.lazy.char8 since that's 8-bit ascii data , input utf-8.

try replacing import

import qualified data.bytestring.lazy bl 

and use bl.readfile read in data (of course actual name doesn't matter, bl idiomatic shorthand lazy bytestring package).

note use data.text handling unicode text, in case aeson api expects binary (i.e. bytestring) representation , handles decoding unicode internally.

edit: i've thought more, don't think problem using char8 after (although point stands not using unicode text in general) not doing conversions string or char (expect c.pack . c.unpack experiment, break multi-byte characters).