diff options
author | Vaclav Zeman <vhaisman@gmail.com> | 2014-02-09 13:16:39 +0100 |
---|---|---|
committer | Vaclav Zeman <vhaisman@gmail.com> | 2014-02-09 13:16:39 +0100 |
commit | 1ba8066f67d1f8624e679e28fd800e6ade2a9511 (patch) | |
tree | 4bdd9c34936f7be77b95aaf4df723980aa675a25 /src/Text/Pandoc/Readers/Markdown.hs | |
parent | 3f0fe345f9aa69d1faf36e6a6f913013f21b3749 (diff) | |
parent | 927b51630852b56f6c8e0aa4a7763108fdb8ec89 (diff) | |
download | pandoc-1ba8066f67d1f8624e679e28fd800e6ade2a9511.tar.gz |
Merge remote-tracking branch 'origin/master' into en-dash-ligature-avoidance.
Diffstat (limited to 'src/Text/Pandoc/Readers/Markdown.hs')
-rw-r--r-- | src/Text/Pandoc/Readers/Markdown.hs | 91 |
1 files changed, 33 insertions, 58 deletions
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) |