c80ffe9 · typeable/pandoc

git.iko.soy/typeable/pandoc

UTF8: Better error message for invalid UTF8.

Read bytestring and use Text's decodeUtf8 instead of using System.IO's hGetContents. This way you get a message saying "invalid UTF-8 stream" instead of "invalid byte sequence." You are also told which byte caused the problem.

commitc80ffe9171391f1a15d0c4b069f4b994c1aef050 authorJohn MacFarlane <fiddlosopher@gmail.com> dateWed 26 Sep 2012, 09:04 (14 years ago) parent885ef20

1 file changed+6−4

compare with master

unifiedsplit

src/Text/Pandoc/UTF8.hs+6−4

src/Text/Pandoc/UTF8.hs+6−4

@@ -52,7 +52,8 @@ import Codec.Binary.UTF8.String (encodeString, decodeString)

5252

5353import System.IO hiding (readFile, writeFile, getContents,

5454 putStr, putStrLn, hPutStr, hPutStrLn, hGetContents)

55−import Prelude hiding (readFile, writeFile, getContents, putStr, putStrLn )

55+import Prelude hiding (readFile, writeFile, getContents, putStr, putStrLn,

56+ catch)

5657import qualified System.IO as IO

5758import qualified Data.ByteString.Char8 as B

5859import qualified Data.ByteString.Lazy as BL

@@ -86,9 +87,10 @@ hPutStrLn :: Handle -> String -> IO ()

8687hPutStrLn h s = hSetEncoding h utf8 >> IO.hPutStrLn h s

8788

8889hGetContents :: Handle -> IO String

89−hGetContents h = hSetEncoding h utf8_bom

90− >> hSetNewlineMode h universalNewlineMode

91− >> IO.hGetContents h

90+hGetContents h = fmap (TL.unpack . TL.decodeUtf8) $ BL.hGetContents h

91+-- hGetContents h = hSetEncoding h utf8_bom

92+-- >> hSetNewlineMode h universalNewlineMode

93+-- >> IO.hGetContents h

9294

9395toString :: B.ByteString -> String

9496toString = T.unpack . T.decodeUtf8With lenientDecode