diff options
Diffstat (limited to 'src/Text/Pandoc/Readers')
| -rw-r--r-- | src/Text/Pandoc/Readers/Markdown.hs | 4 | ||||
| -rw-r--r-- | src/Text/Pandoc/Readers/Org.hs | 70 |
2 files changed, 59 insertions, 15 deletions
diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 02a787670..7a3be8291 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -735,9 +735,9 @@ anyOrderedListStart = try $ do skipNonindentSpaces notFollowedBy $ string "p." >> spaceChar >> digit -- page number res <- do guardDisabled Ext_fancy_lists - many1 digit + start <- many1 digit >>= safeRead char '.' - return (1, DefaultStyle, DefaultDelim) + return (start, DefaultStyle, DefaultDelim) <|> do (num, style, delim) <- anyOrderedListMarker -- if it could be an abbreviated first name, -- insist on more than one space diff --git a/src/Text/Pandoc/Readers/Org.hs b/src/Text/Pandoc/Readers/Org.hs index 5c00a1b27..a6ebf65dc 100644 --- a/src/Text/Pandoc/Readers/Org.hs +++ b/src/Text/Pandoc/Readers/Org.hs @@ -69,7 +69,32 @@ parseOrg = do blocks' <- parseBlocks st <- getState let meta = runF (orgStateMeta' st) st - return $ Pandoc meta $ filter (/= Null) (B.toList $ runF blocks' st) + let removeUnwantedBlocks = dropCommentTrees . filter (/= Null) + return $ Pandoc meta $ removeUnwantedBlocks (B.toList $ runF blocks' st) + +-- | Drop COMMENT headers and the document tree below those headers. +dropCommentTrees :: [Block] -> [Block] +dropCommentTrees [] = [] +dropCommentTrees blks@(b:bs) = + maybe blks (flip dropUntilHeaderAboveLevel bs) $ commentHeaderLevel b + +-- | Return the level of a header starting a comment tree and Nothing +-- otherwise. +commentHeaderLevel :: Block -> Maybe Int +commentHeaderLevel blk = + case blk of + (Header level _ ((Str "COMMENT"):_)) -> Just level + _ -> Nothing + +-- | Drop blocks until a header on or above the given level is seen +dropUntilHeaderAboveLevel :: Int -> [Block] -> [Block] +dropUntilHeaderAboveLevel n = dropWhile (not . isHeaderLevelLowerEq n) + +isHeaderLevelLowerEq :: Int -> Block -> Bool +isHeaderLevelLowerEq n blk = + case blk of + (Header level _ _) -> n >= level + _ -> False -- -- Parser State for Org @@ -828,12 +853,14 @@ list :: OrgParser (F Blocks) list = choice [ definitionList, bulletList, orderedList ] <?> "list" definitionList :: OrgParser (F Blocks) -definitionList = fmap B.definitionList . fmap compactify'DL . sequence - <$> many1 (definitionListItem bulletListStart) +definitionList = try $ do n <- lookAhead (bulletListStart' Nothing) + fmap B.definitionList . fmap compactify'DL . sequence + <$> many1 (definitionListItem $ bulletListStart' (Just n)) bulletList :: OrgParser (F Blocks) -bulletList = fmap B.bulletList . fmap compactify' . sequence - <$> many1 (listItem bulletListStart) +bulletList = try $ do n <- lookAhead (bulletListStart' Nothing) + fmap B.bulletList . fmap compactify' . sequence + <$> many1 (listItem (bulletListStart' $ Just n)) orderedList :: OrgParser (F Blocks) orderedList = fmap B.orderedList . fmap compactify' . sequence @@ -845,10 +872,27 @@ genericListStart listMarker = try $ (+) <$> (length <$> many spaceChar) <*> (length <$> listMarker <* many1 spaceChar) --- parses bullet list start and returns its length (excl. following whitespace) +-- parses bullet list marker. maybe we know the indent level bulletListStart :: OrgParser Int -bulletListStart = genericListStart bulletListMarker - where bulletListMarker = pure <$> oneOf "*-+" +bulletListStart = bulletListStart' Nothing + +bulletListStart' :: Maybe Int -> OrgParser Int +-- returns length of bulletList prefix, inclusive of marker +bulletListStart' Nothing = do ind <- many spaceChar + oneOf bullets + many1 spaceChar + return $ length ind + 1 + -- Unindented lists are legal, but they can't use '*' bullets + -- We return n to maintain compatibility with the generic listItem +bulletListStart' (Just n) = do count (n-1) spaceChar + oneOf validBullets + many1 spaceChar + return n + where validBullets = if n == 1 then noAsterisks else bullets + noAsterisks = filter (/= '*') bullets + +bullets :: String +bullets = "*+-" orderedListStart :: OrgParser Int orderedListStart = genericListStart orderedListMarker @@ -927,7 +971,7 @@ parseInlines = trimInlinesF . mconcat <$> many1 inline -- treat these as potentially non-text when parsing inline: specialChars :: [Char] -specialChars = "\"$'()*+-./:<=>[\\]^_{|}~" +specialChars = "\"$'()*+-,./:<=>[\\]^_{|}~" whitespace :: OrgParser (F Inlines) @@ -1205,10 +1249,10 @@ displayMath = return . B.displayMath <$> choice [ rawMathBetween "\\[" "\\]" ] symbol :: OrgParser (F Inlines) symbol = return . B.str . (: "") <$> (oneOf specialChars >>= updatePositions) - where updatePositions c - | c `elem` emphasisPreChars = c <$ updateLastPreCharPos - | c `elem` emphasisForbiddenBorderChars = c <$ updateLastForbiddenCharPos - | otherwise = return c + where updatePositions c = do + when (c `elem` emphasisPreChars) updateLastPreCharPos + when (c `elem` emphasisForbiddenBorderChars) updateLastForbiddenCharPos + return c emphasisBetween :: Char -> OrgParser (F Inlines) |