Simon Peyton-Jones writes:
> Somehow the LINE pragmas in ParBNF.hs have lost their trailing "#".
> They should say
> {-# LINE 1 "GenericTemplate.hs" #-}
> but actually they say
> {-# LINE 1 "GenericTemplate.hs" -}
This is one problem -- thanks for the hint! --, but there appears to
be another one. After editing these lines, I still get an error:
| $ make
| ghc -iformats -iformats/haskell -iformats/java -iformats/cpp -iformats/c -package
util --make CFTop -o bnfc
| Chasing modules from: CFTop
| Skipping Alex ( ./Alex.hs, ./Alex.o )
| Skipping ErrM ( ./ErrM.hs, ./ErrM.o )
| Skipping LexBNF ( ./LexBNF.hs, ./LexBNF.o )
| Skipping AbsBNF ( ./AbsBNF.hs, ./AbsBNF.o )
| Skipping ParBNF ( ./ParBNF.hs, ./ParBNF.o )
| Skipping RegToFlex ( formats/cpp/RegToFlex.hs, formats/cpp/RegToFlex.o )
| Skipping RegToJLex ( formats/java/RegToJLex.hs, formats/java/RegToJLex.o )
| Skipping RegToAlex ( formats/haskell/RegToAlex.hs,
formats/haskell/RegToAlex.o )
| Skipping Utils ( ./Utils.hs, ./Utils.o )
| Compiling CF ( ./CF.hs, ./CF.o )
| ./CF.hs:406: lexical error
And the code in that block doesn't look incorrect:
badList = isList c &&
not (isCoercion f ||�isNilFun f || isOneFun f || isConsFun f)
Also, ghc-6.0 compiles it fine.
Any idea what this might be?
Peter
_______________________________________________
Cvs-ghc mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/cvs-ghc
