UTF-8 Conversion Functions

[RobertFischer]

It'd be nice to have explicit functions for converting ByteString to and from UTF8, since that's the common use case.

This is the code that I find very useful in my local library:

asUTF8 :: (ToText a, FromText (UTF8 b)) => a -> b
asUTF8 = unUTF8 . fromText . toText
{-# INLINE asUTF8 #-}
{-# SPECIALIZE INLINE asUTF8 :: String-> LazyByteString          #-}
{-# SPECIALIZE INLINE asUTF8 :: StrictText -> LazyByteString     #-}
{-# SPECIALIZE INLINE asUTF8 :: LazyText -> LazyByteString       #-}
{-# SPECIALIZE INLINE asUTF8 :: String-> StrictByteString        #-}
{-# SPECIALIZE INLINE asUTF8 :: StrictText -> StrictByteString   #-}
{-# SPECIALIZE INLINE asUTF8 :: LazyText -> StrictByteString     #-}
{-# SPECIALIZE INLINE asUTF8 :: String-> ShortByteString         #-}
{-# SPECIALIZE INLINE asUTF8 :: StrictText -> ShortByteString    #-}
{-# SPECIALIZE INLINE asUTF8 :: LazyText -> ShortByteString      #-}

toUTF8 :: (DecodeText f (UTF8 a), FromText b) => a -> f b
toUTF8 = decodeConvertText . UTF8
{-# INLINE toUTF8 #-}
{-# SPECIALIZE INLINE toUTF8 :: LazyByteString -> Maybe String       #-}
{-# SPECIALIZE INLINE toUTF8 :: LazyByteString -> Maybe StrictText   #-}
{-# SPECIALIZE INLINE toUTF8 :: LazyByteString -> Maybe LazyText     #-}
{-# SPECIALIZE INLINE toUTF8 :: StrictByteString -> Maybe String     #-}
{-# SPECIALIZE INLINE toUTF8 :: StrictByteString -> Maybe StrictText #-}
{-# SPECIALIZE INLINE toUTF8 :: StrictByteString -> Maybe LazyText   #-}
{-# SPECIALIZE INLINE toUTF8 :: ShortByteString -> Maybe String      #-}
{-# SPECIALIZE INLINE toUTF8 :: ShortByteString -> Maybe StrictText  #-}
{-# SPECIALIZE INLINE toUTF8 :: ShortByteString -> Maybe LazyText    #-}

If this is something you're interested in, let me know, and I'm happy to submit a pull request.

[jxv]

This is useful. Yes, send a PR.

[lexi-lambda]

The names seem wrong to me. toUTF8 seems like it ought to be fromUTF8, since it is converting from the UTF-8 encoding to a textual representation. Similarly, I think asUTF8 ought to be toUTF8.

That aside, I think these functions seem useful. I would be willing to merge them if you submitted a PR.

[RobertFischer]

Happy to change the names. :)