X-Git-Url: http://git.megacz.com/?a=blobdiff_plain;f=compiler%2Fparser%2FLexer.x;h=e9e9c866a329e6fa915094931fb1425405ea92aa;hb=e9a0f76386366ef0f1554e6cf0024a06252c3013;hp=4caca44a70151bdcabcbeabcb5d5a60e31a29987;hpb=35cb95c2119a3d903ecfe388d3a8ef0f4ededfdd;p=ghc-hetmet.git

diff --git a/compiler/parser/Lexer.x b/compiler/parser/Lexer.x
index 4caca44..e9e9c86 100644
--- a/compiler/parser/Lexer.x
+++ b/compiler/parser/Lexer.x
@@ -58,30 +58,30 @@ import Compat.Unicode	( GeneralCategory(..), generalCategory, isPrint, isUpper )
 #endif
 }
 
-$unispace    = \x05
+$unispace    = \x05 -- Trick Alex into handling Unicode. See alexGetChar.
 $whitechar   = [\ \n\r\f\v\xa0 $unispace]
 $white_no_nl = $whitechar # \n
 $tab         = \t
 
 $ascdigit  = 0-9
-$unidigit  = \x03
+$unidigit  = \x03 -- Trick Alex into handling Unicode. See alexGetChar.
 $decdigit  = $ascdigit -- for now, should really be $digit (ToDo)
 $digit     = [$ascdigit $unidigit]
 
 $special   = [\(\)\,\;\[\]\`\{\}]
 $ascsymbol = [\!\#\$\%\&\*\+\.\/\<\=\>\?\@\\\^\|\-\~ \xa1-\xbf \xd7 \xf7]
-$unisymbol = \x04
+$unisymbol = \x04 -- Trick Alex into handling Unicode. See alexGetChar.
 $symbol    = [$ascsymbol $unisymbol] # [$special \_\:\"\']
 
-$unilarge  = \x01
+$unilarge  = \x01 -- Trick Alex into handling Unicode. See alexGetChar.
 $asclarge  = [A-Z \xc0-\xd6 \xd8-\xde]
 $large     = [$asclarge $unilarge]
 
-$unismall  = \x02
+$unismall  = \x02 -- Trick Alex into handling Unicode. See alexGetChar.
 $ascsmall  = [a-z \xdf-\xf6 \xf8-\xff]
 $small     = [$ascsmall $unismall \_]
 
-$unigraphic = \x06
+$unigraphic = \x06 -- Trick Alex into handling Unicode. See alexGetChar.
 $graphic   = [$small $large $symbol $digit $special $unigraphic \:\"\']
 
 $octit	   = 0-7
@@ -699,8 +699,19 @@ notFollowedBy char _ _ _ (AI _ _ buf)
 notFollowedBySymbol _ _ _ (AI _ _ buf)
   = nextCharIs buf (`notElem` "!#$%&*+./<=>?@\\^|-~")
 
+-- We must reject doc comments as being ordinary comments everywhere.
+-- In some cases the doc comment will be selected as the lexeme due to
+-- maximal munch, but not always, because the nested comment rule is
+-- valid in all states, but the doc-comment rules are only valid in
+-- the non-layout states.
 isNormalComment bits _ _ (AI _ _ buf)
-  = nextCharIs buf (/='#')
+  | haddockEnabled bits = notFollowedByDocOrPragma
+  | otherwise           = nextCharIs buf (/='#')
+  where
+    notFollowedByDocOrPragma
+       = not $ spaceAndP buf (`nextCharIs` (`elem` "|^*$#"))
+
+spaceAndP buf p = p buf || nextCharIs buf (==' ') && p (snd (nextChar buf))
 
 haddockDisabledAnd p bits _ _ (AI _ _ buf)
   = if haddockEnabled bits then False else (p buf)
@@ -1408,6 +1419,9 @@ alexGetChar (AI loc ofs s)
 	adj_c 
 	  | c <= '\x06' = non_graphic
 	  | c <= '\xff' = c
+          -- Alex doesn't handle Unicode, so when Unicode
+          -- character is encoutered we output these values
+          -- with the actual character value hidden in the state.
 	  | otherwise = 
 		case generalCategory c of
 		  UppercaseLetter       -> upper