Re: UTF-8 library
Ashley Yakeley wrote [snip]
Text encoded with ISO 8859-1 or UTF-8 is octets. If you want to use CChars, you should then subsequently convert the Word8s into CChars. We were talking about converting CStrings, which are necessarily sequences of CChars. I have to say I do not relish the prospect of replacing the current peekCString interface by three functions which (1) translate a CString/CStringLen into [CChar] (2) translate a [CChar] into a [Word8] (3) translate a [Word8] into a String (and of course the inverse functions to go in the other direction.)
If we want to do this, we certainly need to keep the existing functions since I certainly don't want to have to pass a String through three separate transformations just to make it suitable for C.
On Wed, Aug 07, 2002 at 03:29:33PM +0200, George Russell wrote:
Ashley Yakeley wrote [snip]
Text encoded with ISO 8859-1 or UTF-8 is octets. If you want to use CChars, you should then subsequently convert the Word8s into CChars. We were talking about converting CStrings, which are necessarily sequences of CChars. I have to say I do not relish the prospect of replacing the current peekCString interface by three functions which (1) translate a CString/CStringLen into [CChar] (2) translate a [CChar] into a [Word8] (3) translate a [Word8] into a String (and of course the inverse functions to go in the other direction.) I guess you can avoid (2) or (3) in practice.
If we want to do this, we certainly need to keep the existing functions since I certainly don't want to have to pass a String through three separate transformations just to make it suitable for C. I don't see a problem with supplying a backward compatible withCString function, even it might use the current locale to do the conversion.
Let's just wait till someone actually has a ready-to-criticise library for ghc at hand. Axel.
participants (2)
-
Axel Simon -
George Russell