Unicode lexer issue?

[Simon Marlow]

Bulat Ziganshin wrote:

> at least i prefer to set the default to latin-1 because my files
> contains huge portrion of Russian comments in 1251 codepage, which is
> a chars > 128. so, such defaulting will break my sources and any other
> written with non-english comments. i think that this is unacceptable,
> especially for automatic download/build scenarios of ghc usage

At the moment, the plan for GHC 6.6 is to accept UTF-8 source files 
only.  In fact, the code is already checked in (please try it!).

If UTF-8 only will cause you a problem, please let me know.  If 
necessary, we can implement recoding into UTF-8 before parsing, using 
iconv or something.

Cheers,
	Simon