Modified rawHtmlBlock in HTML reader so it parses </html> and </body> tags.

This allows these tags to be handled correctly in Markdown.
HTML reader now uses rawHtmlBlock', which excludes </html> and </body>,
since these are handled in parseHtml.  (Resolves Issue #38.)


git-svn-id: https://pandoc.googlecode.com/svn/trunk@1152 788f1e2b-df1e-0410-8736-df70ead52e1b

1 files changed, 7 insertions, 2 deletions

diff --git a/Text/Pandoc/Readers/HTML.hs b/Text/Pandoc/Readers/HTML.hs
index 70a071152..d379b373c 100644
--- a/Text/Pandoc/Readers/HTML.hs
+++ b/Text/Pandoc/Readers/HTML.hs
@@ -199,12 +199,16 @@ htmlScript = try $ do
 htmlBlockElement = choice [ htmlScript, htmlComment, xmlDec, definition ]
 
 rawHtmlBlock = try $ do
-  notFollowedBy' (htmlTag "/body" <|> htmlTag "/html")
   body <- htmlBlockElement <|> anyHtmlTag <|> anyHtmlEndTag
   sp <- many space
   state <- getState
   if stateParseRaw state then return (RawHtml (body ++ sp)) else return Null
 
+-- We don't want to parse </body> or </html> as raw HTML, since these
+-- are handled in parseHtml.
+rawHtmlBlock' = do notFollowedBy' (htmlTag "/body" <|> htmlTag "/html")
+                   rawHtmlBlock
+
 -- | Parses an HTML comment.
 htmlComment = try $ do
   string "<!--"
@@ -284,7 +288,8 @@ block = choice [ codeBlock
                , blockQuote
                , para
                , plain
-               , rawHtmlBlock ] <?> "block"
+               , rawHtmlBlock'
+               ] <?> "block"
 
 --
 -- header blocks