diff options
Diffstat (limited to 'src/Text/Pandoc/Readers')
| -rw-r--r-- | src/Text/Pandoc/Readers/DocBook.hs | 11 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/HTML.hs | 26 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/LaTeX.hs | 36 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/Markdown.hs | 91 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/MediaWiki.hs | 52 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/RST.hs | 23 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/Textile.hs | 2 |
7 files changed, 142 insertions, 99 deletions
diff --git a/src/Text/Pandoc/Readers/DocBook.hs b/src/Text/Pandoc/Readers/DocBook.hs index fc29988d5..56cb16b20 100644 --- a/src/Text/Pandoc/Readers/DocBook.hs +++ b/src/Text/Pandoc/Readers/DocBook.hs @@ -12,6 +12,7 @@ import Data.Char (isSpace) import Control.Monad.State import Control.Applicative ((<$>)) import Data.List (intersperse) +import Data.Maybe (fromMaybe) {- @@ -683,7 +684,7 @@ parseBlock (Elem e) = "lowerroman" -> LowerRoman "upperroman" -> UpperRoman _ -> Decimal - let start = maybe 1 id $ + let start = fromMaybe 1 $ (attrValue "override" <$> filterElement (named "listitem") e) >>= safeRead orderedListWith (start,listStyle,DefaultDelim) @@ -779,7 +780,7 @@ parseBlock (Elem e) = caption <- case filterChild isCaption e of Just t -> getInlines t Nothing -> return mempty - let e' = maybe e id $ filterChild (named "tgroup") e + let e' = fromMaybe e $ filterChild (named "tgroup") e let isColspec x = named "colspec" x || named "col" x let colspecs = case filterChild (named "colgroup") e' of Just c -> filterChildren isColspec c @@ -801,12 +802,14 @@ parseBlock (Elem e) = Just "center" -> AlignCenter _ -> AlignDefault let toWidth c = case findAttr (unqual "colwidth") c of - Just w -> maybe 0 id + Just w -> fromMaybe 0 $ safeRead $ '0': filter (\x -> (x >= '0' && x <= '9') || x == '.') w Nothing -> 0 :: Double - let numrows = maximum $ map length bodyrows + let numrows = case bodyrows of + [] -> 0 + xs -> maximum $ map length xs let aligns = case colspecs of [] -> replicate numrows AlignDefault cs -> map toAlignment cs diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index 80279bf61..d1e4d0024 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -76,9 +76,18 @@ pBody :: TagParser [Block] pBody = pInTags "body" block pHead :: TagParser [Block] -pHead = pInTags "head" $ pTitle <|> ([] <$ pAnyTag) +pHead = pInTags "head" $ pTitle <|> pMetaTag <|> ([] <$ pAnyTag) where pTitle = pInTags "title" inline >>= setTitle . normalizeSpaces setTitle t = [] <$ (updateState $ B.setMeta "title" (B.fromList t)) + pMetaTag = do + mt <- pSatisfy (~== TagOpen "meta" []) + let name = fromAttrib "name" mt + if null name + then return [] + else do + let content = fromAttrib "content" mt + updateState $ B.setMeta name (B.text content) + return [] block :: TagParser [Block] block = choice @@ -207,7 +216,7 @@ pHeader = try $ do let bodyTitle = TagOpen tagtype attr ~== TagOpen "h1" [("class","title")] let level = read (drop 1 tagtype) contents <- liftM concat $ manyTill inline (pCloses tagtype <|> eof) - let ident = maybe "" id $ lookup "id" attr + let ident = fromMaybe "" $ lookup "id" attr let classes = maybe [] words $ lookup "class" attr let keyvals = [(k,v) | (k,v) <- attr, k /= "class", k /= "id"] return $ if bodyTitle @@ -257,7 +266,7 @@ pCol = try $ do skipMany pBlank return $ case lookup "width" attribs of Just x | not (null x) && last x == '%' -> - maybe 0.0 id $ safeRead ('0':'.':init x) + fromMaybe 0.0 $ safeRead ('0':'.':init x) _ -> 0.0 pColgroup :: TagParser [Double] @@ -467,7 +476,13 @@ pBlank = try $ do pTagContents :: Parser [Char] ParserState Inline pTagContents = - pStr <|> pSpace <|> smartPunctuation pTagContents <|> pSymbol <|> pBad + Math DisplayMath `fmap` mathDisplay + <|> Math InlineMath `fmap` mathInline + <|> pStr + <|> pSpace + <|> smartPunctuation pTagContents + <|> pSymbol + <|> pBad pStr :: Parser [Char] ParserState Inline pStr = do @@ -482,6 +497,7 @@ isSpecial '"' = True isSpecial '\'' = True isSpecial '.' = True isSpecial '-' = True +isSpecial '$' = True isSpecial '\8216' = True isSpecial '\8217' = True isSpecial '\8220' = True @@ -557,7 +573,7 @@ blockHtmlTags = ["address", "article", "aside", "blockquote", "body", "button", "noframes", "noscript", "object", "ol", "output", "p", "pre", "progress", "section", "table", "tbody", "textarea", "thead", "tfoot", "ul", "dd", "dt", "frameset", "li", "tbody", "td", "tfoot", - "th", "thead", "tr", "script", "style", "video"] + "th", "thead", "tr", "script", "style", "svg", "video"] -- We want to allow raw docbook in markdown documents, so we -- include docbook block tags here too. diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 681dcb077..51271edc5 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -38,7 +38,8 @@ import Text.Pandoc.Definition import Text.Pandoc.Walk import Text.Pandoc.Shared import Text.Pandoc.Options -import Text.Pandoc.Parsing hiding ((<|>), many, optional, space) +import Text.Pandoc.Parsing hiding ((<|>), many, optional, space, + mathDisplay, mathInline) import qualified Text.Pandoc.UTF8 as UTF8 import Data.Char ( chr, ord ) import Control.Monad @@ -163,13 +164,25 @@ mathChars = concat <$> <|> (\c -> ['\\',c]) <$> (try $ char '\\' *> anyChar) ) +quoted' :: (Inlines -> Inlines) -> LP String -> LP () -> LP Inlines +quoted' f starter ender = do + startchs <- starter + try ((f . mconcat) <$> manyTill inline ender) <|> lit startchs + double_quote :: LP Inlines -double_quote = (doubleQuoted . mconcat) <$> - (try $ string "``" *> manyTill inline (try $ string "''")) +double_quote = + ( quoted' doubleQuoted (try $ string "``") (void $ try $ string "''") + <|> quoted' doubleQuoted (string "“") (void $ char '”') + -- the following is used by babel for localized quotes: + <|> quoted' doubleQuoted (try $ string "\"`") (void $ try $ string "\"'") + <|> quoted' doubleQuoted (string "\"") (void $ char '"') + ) single_quote :: LP Inlines -single_quote = (singleQuoted . mconcat) <$> - (try $ char '`' *> manyTill inline (try $ char '\'' >> notFollowedBy letter)) +single_quote = + ( quoted' singleQuoted (string "`") (try $ char '\'' >> notFollowedBy letter) + <|> quoted' singleQuoted (string "‘") (try $ char '’' >> notFollowedBy letter) + ) inline :: LP Inlines inline = (mempty <$ comment) @@ -181,10 +194,10 @@ inline = (mempty <$ comment) ((char '-') *> option (str "–") (str "—" <$ char '-'))) <|> double_quote <|> single_quote - <|> (str "“" <$ try (string "``")) -- nb. {``} won't be caught by double_quote <|> (str "”" <$ try (string "''")) - <|> (str "‘" <$ char '`') -- nb. {`} won't be caught by single_quote + <|> (str "”" <$ char '”') <|> (str "’" <$ char '\'') + <|> (str "’" <$ char '’') <|> (str "\160" <$ char '~') <|> (mathDisplay $ string "$$" *> mathChars <* string "$$") <|> (mathInline $ char '$' *> mathChars <* char '$') @@ -373,6 +386,7 @@ inlineCommands = M.fromList $ , ("backslash", lit "\\") , ("slash", lit "/") , ("textbf", strong <$> tok) + , ("textnormal", spanWith ("",["nodecor"],[]) <$> tok) , ("ldots", lit "…") , ("dots", lit "…") , ("mdots", lit "…") @@ -443,6 +457,7 @@ inlineCommands = M.fromList $ , ("footnote", (note . mconcat) <$> (char '{' *> manyTill block (char '}'))) , ("verb", doverb) , ("lstinline", doverb) + , ("Verb", doverb) , ("texttt", (code . stringify . toList) <$> tok) , ("url", (unescapeURL <$> braced) >>= \url -> pure (link url "" (str url))) @@ -754,7 +769,7 @@ inlineText :: LP Inlines inlineText = str <$> many1 inlineChar inlineChar :: LP Char -inlineChar = noneOf "\\$%^_&~#{}^'`-[] \t\n" +inlineChar = noneOf "\\$%^_&~#{}^'`\"‘’“”-[] \t\n" environment :: LP Blocks environment = do @@ -859,9 +874,8 @@ verbatimEnv = do (_,r) <- withRaw $ do controlSeq "begin" name <- braced - guard $ name == "verbatim" || name == "Verbatim" || - name == "lstlisting" || name == "minted" || - name == "alltt" + guard $ name `elem` ["verbatim", "Verbatim", "lstlisting", + "minted", "alltt"] verbEnv name rest <- getInput return (r,rest) diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index ea49d8c1d..8a41cef49 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -215,10 +215,10 @@ pandocTitleBlock = try $ do author' <- author date' <- date return $ - ( if B.isNull title' then id else B.setMeta "title" title' - . if null author' then id else B.setMeta "author" author' - . if B.isNull date' then id else B.setMeta "date" date' ) - nullMeta + (if B.isNull title' then id else B.setMeta "title" title') + . (if null author' then id else B.setMeta "author" author') + . (if B.isNull date' then id else B.setMeta "date" date') + $ nullMeta updateState $ \st -> st{ stateMeta' = stateMeta' st <> meta' } yamlMetaBlock :: MarkdownParser (F Blocks) @@ -730,9 +730,14 @@ listStart = bulletListStart <|> (anyOrderedListStart >> return ()) listLine :: MarkdownParser String listLine = try $ do notFollowedBy' (do indentSpaces - many (spaceChar) + many spaceChar listStart) - chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 anyChar) newline + notFollowedBy' $ htmlTag (~== TagClose "div") + chunks <- manyTill + ( many1 (satisfy $ \c -> c /= '\n' && c /= '<') + <|> liftM snd (htmlTag isCommentTag) + <|> count 1 anyChar + ) newline return $ concat chunks -- parse raw text for one list item, excluding start marker and continuations @@ -759,6 +764,7 @@ listContinuationLine :: MarkdownParser String listContinuationLine = try $ do notFollowedBy blankline notFollowedBy' listStart + notFollowedBy' $ htmlTag (~== TagClose "div") optional indentSpaces result <- anyLine return $ result ++ "\n" @@ -783,8 +789,8 @@ listItem start = try $ do orderedList :: MarkdownParser (F Blocks) orderedList = try $ do (start, style, delim) <- lookAhead anyOrderedListStart - unless ((style == DefaultStyle || style == Decimal || style == Example) && - (delim == DefaultDelim || delim == Period)) $ + unless (style `elem` [DefaultStyle, Decimal, Example] && + delim `elem` [DefaultDelim, Period]) $ guardEnabled Ext_fancy_lists when (style == Example) $ guardEnabled Ext_example_lists items <- fmap sequence $ many1 $ listItem @@ -896,7 +902,9 @@ plain = fmap B.plain . trimInlinesF . mconcat <$> many1 inline -- htmlElement :: MarkdownParser String -htmlElement = strictHtmlBlock <|> liftM snd (htmlTag isBlockTag) +htmlElement = rawVerbatimBlock + <|> strictHtmlBlock + <|> liftM snd (htmlTag isBlockTag) htmlBlock :: MarkdownParser (F Blocks) htmlBlock = do @@ -917,8 +925,8 @@ strictHtmlBlock = htmlInBalanced (not . isInlineTag) rawVerbatimBlock :: MarkdownParser String rawVerbatimBlock = try $ do - (TagOpen tag _, open) <- htmlTag (tagOpen (\t -> - t == "pre" || t == "style" || t == "script") + (TagOpen tag _, open) <- htmlTag (tagOpen (flip elem + ["pre", "style", "script"]) (const True)) contents <- manyTill anyChar (htmlTag (~== TagClose tag)) return $ open ++ contents ++ renderTags [TagClose tag] @@ -1118,12 +1126,12 @@ multilineTableHeader headless = try $ do then liftM (map (:[]) . tail . splitStringByIndices (init indices)) $ lookAhead anyLine else return $ transpose $ map - (\ln -> tail $ splitStringByIndices (init indices) ln) + (tail . splitStringByIndices (init indices)) rawContent let aligns = zipWith alignType rawHeadsList lengths let rawHeads = if headless then replicate (length dashes) "" - else map unwords rawHeadsList + else map (unlines . map trim) rawHeadsList heads <- fmap sequence $ mapM (parseFromString (mconcat <$> many plain)) $ map trim rawHeads @@ -1180,7 +1188,7 @@ gridTableHeader headless = try $ do -- RST does not have a notion of alignments let rawHeads = if headless then replicate (length dashes) "" - else map unwords $ transpose + else map (unlines . map trim) $ transpose $ map (gridTableSplitLine indices) rawContent heads <- fmap sequence $ mapM (parseFromString parseBlocks . trim) rawHeads return (heads, aligns, indices) @@ -1406,39 +1414,6 @@ math :: MarkdownParser (F Inlines) math = (return . B.displayMath <$> (mathDisplay >>= applyMacros')) <|> (return . B.math <$> (mathInline >>= applyMacros')) -mathDisplay :: MarkdownParser String -mathDisplay = - (guardEnabled Ext_tex_math_dollars >> mathDisplayWith "$$" "$$") - <|> (guardEnabled Ext_tex_math_single_backslash >> - mathDisplayWith "\\[" "\\]") - <|> (guardEnabled Ext_tex_math_double_backslash >> - mathDisplayWith "\\\\[" "\\\\]") - -mathDisplayWith :: String -> String -> MarkdownParser String -mathDisplayWith op cl = try $ do - string op - many1Till (noneOf "\n" <|> (newline >>~ notFollowedBy' blankline)) (try $ string cl) - -mathInline :: MarkdownParser String -mathInline = - (guardEnabled Ext_tex_math_dollars >> mathInlineWith "$" "$") - <|> (guardEnabled Ext_tex_math_single_backslash >> - mathInlineWith "\\(" "\\)") - <|> (guardEnabled Ext_tex_math_double_backslash >> - mathInlineWith "\\\\(" "\\\\)") - -mathInlineWith :: String -> String -> MarkdownParser String -mathInlineWith op cl = try $ do - string op - notFollowedBy space - words' <- many1Till (count 1 (noneOf "\n\\") - <|> (char '\\' >> anyChar >>= \c -> return ['\\',c]) - <|> count 1 newline <* notFollowedBy' blankline - *> return " ") - (try $ string cl) - notFollowedBy digit -- to prevent capture of $5 - return $ concat words' - -- Parses material enclosed in *s, **s, _s, or __s. -- Designed to avoid backtracking. enclosure :: Char @@ -1455,6 +1430,7 @@ enclosure c = do -- Parse inlines til you hit one c or a sequence of two cs. -- If one c, emit emph and then parse two. -- If two cs, emit strong and then parse one. +-- Otherwise, emit ccc then the results. three :: Char -> MarkdownParser (F Inlines) three c = do contents <- mconcat <$> many (notFollowedBy (char c) >> inline) @@ -1479,7 +1455,7 @@ one c prefix' = do contents <- mconcat <$> many ( (notFollowedBy (char c) >> inline) <|> try (string [c,c] >> notFollowedBy (char c) >> - two c prefix') ) + two c mempty) ) (char c >> return (B.emph <$> (prefix' <> contents))) <|> return (return (B.str [c]) <> (prefix' <> contents)) @@ -1564,16 +1540,14 @@ endline :: MarkdownParser (F Inlines) endline = try $ do newline notFollowedBy blankline + -- parse potential list-starts differently if in a list: + st <- getState + when (stateParserContext st == ListItemState) $ notFollowedBy listStart guardDisabled Ext_lists_without_preceding_blankline <|> notFollowedBy listStart guardEnabled Ext_blank_before_blockquote <|> notFollowedBy emailBlockQuoteStart guardEnabled Ext_blank_before_header <|> notFollowedBy (char '#') -- atx header guardEnabled Ext_backtick_code_blocks >> notFollowedBy (() <$ (lookAhead (char '`') >> codeBlockFenced)) - -- parse potential list-starts differently if in a list: - st <- getState - when (stateParserContext st == ListItemState) $ do - notFollowedBy' bulletListStart - notFollowedBy' anyOrderedListStart (guardEnabled Ext_hard_line_breaks >> return (return B.linebreak)) <|> (guardEnabled Ext_ignore_line_breaks >> return mempty) <|> (return $ return B.space) @@ -1746,7 +1720,7 @@ spanHtml = try $ do guardEnabled Ext_markdown_in_html_blocks (TagOpen _ attrs, _) <- htmlTag (~== TagOpen "span" []) contents <- mconcat <$> manyTill inline (htmlTag (~== TagClose "span")) - let ident = maybe "" id $ lookup "id" attrs + let ident = fromMaybe "" $ lookup "id" attrs let classes = maybe [] words $ lookup "class" attrs let keyvals = [(k,v) | (k,v) <- attrs, k /= "id" && k /= "class"] return $ B.spanWith (ident, classes, keyvals) <$> contents @@ -1756,7 +1730,7 @@ divHtml = try $ do guardEnabled Ext_markdown_in_html_blocks (TagOpen _ attrs, _) <- htmlTag (~== TagOpen "div" []) contents <- mconcat <$> manyTill block (htmlTag (~== TagClose "div")) - let ident = maybe "" id $ lookup "id" attrs + let ident = fromMaybe "" $ lookup "id" attrs let classes = maybe [] words $ lookup "class" attrs let keyvals = [(k,v) | (k,v) <- attrs, k /= "id" && k /= "class"] return $ B.divWith (ident, classes, keyvals) <$> contents @@ -1835,9 +1809,10 @@ citeKey = try $ do guard $ lastStrPos /= Just pos suppress_author <- option False (char '-' >> return True) char '@' - first <- letter - let internal p = try $ p >>~ lookAhead (letter <|> digit) - rest <- many $ letter <|> digit <|> internal (oneOf ":.#$%&-_+?<>~/") + first <- letter <|> char '_' + let regchar = satisfy (\c -> isAlphaNum c || c == '_') + let internal p = try $ p >>~ lookAhead regchar + rest <- many $ regchar <|> internal (oneOf ":.#$%&-+?<>~/") let key = first:rest return (suppress_author, key) diff --git a/src/Text/Pandoc/Readers/MediaWiki.hs b/src/Text/Pandoc/Readers/MediaWiki.hs index 136701bd0..794890eb6 100644 --- a/src/Text/Pandoc/Readers/MediaWiki.hs +++ b/src/Text/Pandoc/Readers/MediaWiki.hs @@ -1,4 +1,5 @@ -{-# LANGUAGE RelaxedPolyRec #-} -- needed for inlinesBetween on GHC < 7 +{-# LANGUAGE RelaxedPolyRec, FlexibleInstances, TypeSynonymInstances #-} +-- RelaxedPolyRec needed for inlinesBetween on GHC < 7 {- Copyright (C) 2012 John MacFarlane <jgm@berkeley.edu> @@ -51,7 +52,9 @@ import Data.List (intersperse, intercalate, isPrefixOf ) import Text.HTML.TagSoup import Data.Sequence (viewl, ViewL(..), (<|)) import qualified Data.Foldable as F +import qualified Data.Map as M import Data.Char (isDigit, isSpace) +import Data.Maybe (fromMaybe) -- | Read mediawiki from an input string and return a Pandoc document. readMediaWiki :: ReaderOptions -- ^ Reader options @@ -62,6 +65,8 @@ readMediaWiki opts s = , mwMaxNestingLevel = 4 , mwNextLinkNumber = 1 , mwCategoryLinks = [] + , mwHeaderMap = M.empty + , mwIdentifierList = [] } "source" (s ++ "\n") of Left err' -> error $ "\nError:\n" ++ show err' @@ -71,10 +76,23 @@ data MWState = MWState { mwOptions :: ReaderOptions , mwMaxNestingLevel :: Int , mwNextLinkNumber :: Int , mwCategoryLinks :: [Inlines] + , mwHeaderMap :: M.Map Inlines String + , mwIdentifierList :: [String] } type MWParser = Parser [Char] MWState +instance HasReaderOptions MWParser where + askReaderOption f = (f . mwOptions) `fmap` getState + +instance HasHeaderMap MWParser where + getHeaderMap = fmap mwHeaderMap getState + putHeaderMap hm = updateState $ \st -> st{ mwHeaderMap = hm } + +instance HasIdentifierList MWParser where + getIdentifierList = fmap mwIdentifierList getState + putIdentifierList l = updateState $ \st -> st{ mwIdentifierList = l } + -- -- auxiliary functions -- @@ -131,9 +149,16 @@ inlinesInTags tag = try $ do blocksInTags :: String -> MWParser Blocks blocksInTags tag = try $ do (_,raw) <- htmlTag (~== TagOpen tag []) + let closer = if tag == "li" + then htmlTag (~== TagClose "li") + <|> lookAhead ( + htmlTag (~== TagOpen "li" []) + <|> htmlTag (~== TagClose "ol") + <|> htmlTag (~== TagClose "ul")) + else htmlTag (~== TagClose tag) if '/' `elem` raw -- self-closing tag then return mempty - else mconcat <$> manyTill block (htmlTag (~== TagClose tag)) + else mconcat <$> manyTill block closer charsInTags :: String -> MWParser [Char] charsInTags tag = try $ do @@ -187,7 +212,7 @@ table = do tableStart styles <- option [] parseAttrs <* blankline let tableWidth = case lookup "width" styles of - Just w -> maybe 1.0 id $ parseWidth w + Just w -> fromMaybe 1.0 $ parseWidth w Nothing -> 1.0 caption <- option mempty tableCaption optional rowsep @@ -268,7 +293,7 @@ tableCell = try $ do Just "center" -> AlignCenter _ -> AlignDefault let width = case lookup "width" attrs of - Just xs -> maybe 0.0 id $ parseWidth xs + Just xs -> fromMaybe 0.0 $ parseWidth xs Nothing -> 0.0 return ((align, width), bs) @@ -351,7 +376,8 @@ header = try $ do let lev = length eqs guard $ lev <= 6 contents <- trimInlines . mconcat <$> manyTill inline (count lev $ char '=') - return $ B.header lev contents + attr <- registerHeader nullAttr contents + return $ B.headerWith attr lev contents bulletList :: MWParser Blocks bulletList = B.bulletList <$> @@ -362,15 +388,13 @@ bulletList = B.bulletList <$> orderedList :: MWParser Blocks orderedList = (B.orderedList <$> many1 (listItem '#')) - <|> (B.orderedList <$> (htmlTag (~== TagOpen "ul" []) *> spaces *> - many (listItem '#' <|> li) <* - optional (htmlTag (~== TagClose "ul")))) - <|> do (tag,_) <- htmlTag (~== TagOpen "ol" []) - spaces - items <- many (listItem '#' <|> li) - optional (htmlTag (~== TagClose "ol")) - let start = maybe 1 id $ safeRead $ fromAttrib "start" tag - return $ B.orderedListWith (start, DefaultStyle, DefaultDelim) items + <|> try + (do (tag,_) <- htmlTag (~== TagOpen "ol" []) + spaces + items <- many (listItem '#' <|> li) + optional (htmlTag (~== TagClose "ol")) + let start = fromMaybe 1 $ safeRead $ fromAttrib "start" tag + return $ B.orderedListWith (start, DefaultStyle, DefaultDelim) items) definitionList :: MWParser Blocks definitionList = B.definitionList <$> many1 defListItem diff --git a/src/Text/Pandoc/Readers/RST.hs b/src/Text/Pandoc/Readers/RST.hs index 32893128a..c12a1493a 100644 --- a/src/Text/Pandoc/Readers/RST.hs +++ b/src/Text/Pandoc/Readers/RST.hs @@ -347,14 +347,25 @@ lhsCodeBlock = try $ do getPosition >>= guard . (==1) . sourceColumn guardEnabled Ext_literate_haskell optional codeBlockStart - lns <- many1 birdTrackLine - -- if (as is normal) there is always a space after >, drop it - let lns' = if all (\ln -> null ln || take 1 ln == " ") lns - then map (drop 1) lns - else lns + lns <- latexCodeBlock <|> birdCodeBlock blanklines return $ B.codeBlockWith ("", ["sourceCode", "literate", "haskell"], []) - $ intercalate "\n" lns' + $ intercalate "\n" lns + +latexCodeBlock :: Parser [Char] st [[Char]] +latexCodeBlock = try $ do + try (latexBlockLine "\\begin{code}") + many1Till anyLine (try $ latexBlockLine "\\end{code}") + where + latexBlockLine s = skipMany spaceChar >> string s >> blankline + +birdCodeBlock :: Parser [Char] st [[Char]] +birdCodeBlock = filterSpace <$> many1 birdTrackLine + where filterSpace lns = + -- if (as is normal) there is always a space after >, drop it + if all (\ln -> null ln || take 1 ln == " ") lns + then map (drop 1) lns + else lns birdTrackLine :: Parser [Char] st [Char] birdTrackLine = char '>' >> anyLine diff --git a/src/Text/Pandoc/Readers/Textile.hs b/src/Text/Pandoc/Readers/Textile.hs index 23e07f621..93658cdea 100644 --- a/src/Text/Pandoc/Readers/Textile.hs +++ b/src/Text/Pandoc/Readers/Textile.hs @@ -594,7 +594,7 @@ surrounded border = enclosed (border *> notFollowedBy (oneOf " \t\n\r")) (try bo simpleInline :: Parser [Char] ParserState t -- ^ surrounding parser -> ([Inline] -> Inline) -- ^ Inline constructor -> Parser [Char] ParserState Inline -- ^ content parser (to be used repeatedly) -simpleInline border construct = surrounded border (inlineWithAttribute) >>= +simpleInline border construct = surrounded border inlineWithAttribute >>= return . construct . normalizeSpaces where inlineWithAttribute = (try $ optional attributes) >> inline |