September 2004 - Nhc-bugs

Trouble building nhc98 for AIX
by John Goerzen 04 Oct '04

04 Oct '04

Hello, I have nhc98 1.16 plus the gcc3.3 patch from the website and am trying to build it on AIX 5.1L. First, every test for including inttypes.h instead of stdint.h should include _AIX in the list of systems that use inttypes.h. (Look in mark.h, HsFFI.h). Now then -- I have no existing Haskell support on this machine so I'm bootstrapping from GCC 3.3.2. I am getting this problem: make[1]: Leaving directory `/home/jgoerzen/programs/nhc98-1.16/src/prelude' touch targets/powerpc-AIX-5.1.0.0/prelude-gcc touch targets/powerpc-AIX-5.1.0.0/prelude script/nhc98 -o lib/powerpc-AIX-5.1.0.0/hmake-PRAGMA script/hmake-PRAGMA.hc touch targets/powerpc-AIX-5.1.0.0/pragma-gcc make: *** No rule to make target `src/compiler98/*.hc', needed by `targets/powerpc-AIX-5.1.0.0/compiler-gcc'. Stop. jgoerzen@mfgpro:~/programs/nhc98-1.16$ make make: *** No rule to make target `src/compiler98/*.hc', needed by `targets/powerpc-AIX-5.1.0.0/compiler-gcc'. Stop. So now I am really stuck. Suggestions welcome :-) -- John Goerzen

3 10

Re: [nhc-bugs] another unicode bug
by Antonio Regidor García 30 Sep '04

30 Sep '04

Thanks, I will look at the nhc code and try to implement it (slowly, anyway...). Regards. Antonio Regidor García > The Haskell'98 Report says: > > Haskell uses the Unicode [11] character set. However, source > programs are currently biased toward the ASCII character set used > in earlier versions of Haskell. > > I'm afraid it is still true that all the compiler implementations > accept only 8-bit characters in source files. You can notate a > larger > character with a hex escape e.g. '\x10d5', but often this is > truncated > to the lower 8-bits on output. > > The use of Unicode in Haskell remains a hope, not a reality. > > Regards, > Malcolm ______________________________________________ Renovamos el Correo Yahoo!: ¡100 MB GRATIS! Nuevos servicios, más seguridad http://correo.yahoo.es

1 0

another unicode bug
by Antonio Regidor García 30 Sep '04

30 Sep '04

I can't compile this function change :: Char -> Char change 'a' = 'b' change x = x if I replace 'b' with a non-ascii character. The compiler complaints "Character literal with length>1". Antonio Regidor García ______________________________________________ Renovamos el Correo Yahoo!: ¡100 MB GRATIS! Nuevos servicios, más seguridad http://correo.yahoo.es

2 1

web page bug
by Antonio Regidor García 30 Sep '04

30 Sep '04

The links at: http://www.haskell.org/nhc98/bugs.html doesn't work. Cheers, Antonio Regidor García ______________________________________________ Renovamos el Correo Yahoo!: ¡100 MB GRATIS! Nuevos servicios, más seguridad http://correo.yahoo.es

2 1

non ASCII characters
by Antonio Regidor García 30 Sep '04

30 Sep '04

When executing this program: module Main where main :: IO () main = do putStrLn aNonASCIIstring x <- getChar putStrLn anotherNonASCIIstring putChar x If the 2 strings contain only ASCII characters there is no problem, but if they contain some non-ASCII characters (I used UTF-8) they are displayed incorrectly. OTOH, If I input a non-ASCII character for x, it is displayed correctly. Also, I tried to compile it with GHC and Helium, but they produce a compile error ("lexical error in string/character literal" and "invalid character literal" resp.), under GNU/Linux and Windows. So I don't think this is a problem of my locale (UTF-8) or NHC configuration. I tried Latin-1 suplement and CJK unified ideographs, with this result. Antonio Regidor García ______________________________________________ Renovamos el Correo Yahoo!: ¡100 MB GRATIS! Nuevos servicios, más seguridad http://correo.yahoo.es

1 0