4e29433 · typeable/pandoc

git.iko.soy/typeable/pandoc

on master help sign in

MediaWiki reader: Improvements to list parsing and HTML tag handling.

4e29433John MacFarlane14 years ago

src/Text/Pandoc/Readers/MediaWiki.hs+16−2

compare with master

unifiedsplit

@@ -34,7 +34,6 @@ _ tests for lists

34_ support HTML lists

35_ support list style attributes and start values in ol lists, also

36 value attribute on li

37−_ support <p> tags in lists (and out?)

38_ support :, ::, etc. for indent (treat as list continuation paras?)

37_ support :, ::, etc. for indent (treat as list continuation paras?)

39_ support preformatted text (lines starting with space)

38_ support preformatted text (lines starting with space)

40_ support preformatted text blocks

39_ support preformatted text blocks

@@ -69,10 +68,12 @@ import Text.Pandoc.XML ( fromEntities )

69import Text.Pandoc.Parsing

68import Text.Pandoc.Parsing

70import Text.Pandoc.Shared ( stripTrailingNewlines )

69import Text.Pandoc.Shared ( stripTrailingNewlines )

71import Data.Monoid (mconcat, mempty)

70import Data.Monoid (mconcat, mempty)

71+import qualified Data.Foldable as F

72import Control.Applicative ((<$>), (<*), (*>), (<$))

73import Control.Monad

74import Data.List (intersperse)

75import Text.HTML.TagSoup

76+import Data.Sequence (viewl, ViewL(..), (<|))

77-- | Read mediawiki from an input string and return a Pandoc document.

78-- | Read mediawiki from an input string and return a Pandoc document.

78readMediaWiki :: ReaderOptions -- ^ Reader options

79readMediaWiki :: ReaderOptions -- ^ Reader options

@@ -140,11 +141,20 @@ block = header

140 <|> haskell

141 <|> haskell

141 <|> mempty <$ skipMany1 blankline

142 <|> mempty <$ skipMany1 blankline

142 <|> mempty <$ try (spaces *> htmlComment)

143 <|> mempty <$ try (spaces *> htmlComment)

144+ <|> pTag

145+ <|> blockHtml

143 <|> para

146 <|> para

144

147

145para :: MWParser Blocks

148para :: MWParser Blocks

146para = B.para . trimInlines . mconcat <$> many1 inline

149para = B.para . trimInlines . mconcat <$> many1 inline

147

150

151+-- We can just skip pTags, as contents will be treated as paragraphs

152+pTag :: MWParser Blocks

153+pTag = mempty <$ (htmlTag (\t -> t ~== TagOpen "p" [] || t ~== TagClose "p"))

154+

155+blockHtml :: MWParser Blocks

156+blockHtml = (B.rawBlock "html" . snd <$> htmlTag isBlockTag)

157+

148hrule :: MWParser Blocks

158hrule :: MWParser Blocks

149hrule = B.horizontalRule <$ try (string "----" *> many (char '-') *> newline)

159hrule = B.horizontalRule <$ try (string "----" *> many (char '-') *> newline)

150

160

@@ -223,7 +233,11 @@ listItem' c = try $ do

223 first <- manyTill anyChar newline

233 first <- manyTill anyChar newline

224 rest <- many (try $ char c *> lookAhead listStartChar *>

234 rest <- many (try $ char c *> lookAhead listStartChar *>

225 manyTill anyChar newline)

235 manyTill anyChar newline)

226− parseFromString (mconcat <$> many1 block) $ unlines $ first : rest

236+ contents <- parseFromString (mconcat <$> many1 block)

237+ $ unlines $ first : rest

238+ case viewl (B.unMany contents) of

239+ (Para xs) :< rest -> return $ B.Many $ (Plain xs) <| rest

240+ _ -> return contents

227

241

228--

242--

229-- inline parsers

243-- inline parsers