| Safe Haskell | Safe-Inferred |
|---|---|
| Language | Haskell2010 |
System.OsString.Windows
Synopsis
- data WindowsString
- data WindowsChar
- encodeUtf :: MonadThrow m => String -> m WindowsString
- encodeWith :: TextEncoding -> String -> Either EncodingException WindowsString
- encodeFS :: String -> IO WindowsString
- fromBytes :: MonadThrow m => ByteString -> m WindowsString
- pstr :: QuasiQuoter
- pack :: [WindowsChar] -> WindowsString
- decodeUtf :: MonadThrow m => WindowsString -> m String
- decodeWith :: TextEncoding -> WindowsString -> Either EncodingException String
- decodeFS :: WindowsString -> IO String
- unpack :: WindowsString -> [WindowsChar]
- unsafeFromChar :: Char -> WindowsChar
- toChar :: WindowsChar -> Char
Types
data WindowsString Source #
Commonly used windows string as wide character bytes.
Instances
data WindowsChar Source #
Instances
String construction
encodeUtf :: MonadThrow m => String -> m WindowsString Source #
Partial unicode friendly encoding.
This encodes as UTF16-LE (strictly), which is a pretty good guess.
Throws an EncodingException if encoding fails.
encodeWith :: TextEncoding -> String -> Either EncodingException WindowsString Source #
Encode a String with the specified encoding.
encodeFS :: String -> IO WindowsString Source #
This mimics the behavior of the base library when doing filesystem operations, which does permissive UTF-16 encoding, where coding errors generate Chars in the surrogate range.
The reason this is in IO is because it unifies with the Posix counterpart,
which does require IO. This is safe to unsafePerformIO/unsafeDupablePerformIO.
fromBytes :: MonadThrow m => ByteString -> m WindowsString Source #
Constructs a platform string from a ByteString.
This ensures valid UCS-2LE. Note that this doesn't expand Word8 to Word16 on windows, so you may get invalid UTF-16.
Throws EncodingException on invalid UCS-2LE (although unlikely).
pstr :: QuasiQuoter Source #
QuasiQuote a WindowsString. This accepts Unicode characters
and encodes as UTF-16LE on windows.
pack :: [WindowsChar] -> WindowsString Source #
Pack a list of platform words to a platform string.
Note that using this in conjunction with unsafeFromChar to
convert from [Char] to platform string is probably not what
you want, because it will truncate unicode code points.
String deconstruction
decodeUtf :: MonadThrow m => WindowsString -> m String Source #
Partial unicode friendly decoding.
This decodes as UTF16-LE (strictly), which is a pretty good.
Throws a EncodingException if decoding fails.
decodeWith :: TextEncoding -> WindowsString -> Either EncodingException String Source #
Decode a WindowsString with the specified encoding.
The String is forced into memory to catch all exceptions.
decodeFS :: WindowsString -> IO String Source #
Like decodeUtf, except this mimics the behavior of the base library when doing filesystem
operations, which does permissive UTF-16 encoding, where coding errors generate
Chars in the surrogate range.
The reason this is in IO is because it unifies with the Posix counterpart,
which does require IO. unsafePerformIO/unsafeDupablePerformIO are safe, however.
unpack :: WindowsString -> [WindowsChar] Source #
Unpack a platform string to a list of platform words.
Word construction
unsafeFromChar :: Char -> WindowsChar Source #
Truncates to 2 octets.
Word deconstruction
toChar :: WindowsChar -> Char Source #
Converts back to a unicode codepoint (total).