@@ -1,3 +1,7 @@
1+{-# LANGUAGE BangPatterns #-}
3+{-# LANGUAGE FlexibleInstances #-}
4+{-# LANGUAGE MultiParamTypeClasses #-}
1{-# LANGUAGE OverloadedStrings #-}
5{-# LANGUAGE OverloadedStrings #-}
2{-# LANGUAGE PatternGuards #-}
6{-# LANGUAGE PatternGuards #-}
3{-# LANGUAGE ScopedTypeVariables #-}
7{-# LANGUAGE ScopedTypeVariables #-}
@@ -18,58 +22,50 @@ module Text.Pandoc.Readers.LaTeX ( readLaTeX,
18 applyMacros,
22 applyMacros,
19 rawLaTeXInline,
23 rawLaTeXInline,
20 rawLaTeXBlock,
24 rawLaTeXBlock,
21− inlineCommand
25+ inlineCommand,
24import Control.Applicative (many, optional, (<|>))
30import Control.Applicative (many, optional, (<|>))
25import Control.Monad
31import Control.Monad
26import Control.Monad.Except (throwError)
32import Control.Monad.Except (throwError)
27−import Data.Char (isDigit, isLetter, isAlphaNum, toUpper, chr)
33+import Data.Char (isDigit, isLetter, toUpper, chr)
28import Data.Default
34import Data.Default
35+import Data.Functor (($>))
29import Data.List (intercalate)
36import Data.List (intercalate)
30import qualified Data.Map as M
37import qualified Data.Map as M
31import Data.Maybe (fromMaybe, maybeToList)
38import Data.Maybe (fromMaybe, maybeToList)
32import qualified Data.Set as Set
39import qualified Data.Set as Set
40+import qualified Data.Sequence as Seq
33import Data.Text (Text)
41import Data.Text (Text)
34import qualified Data.Text as T
42import qualified Data.Text as T
35−import Skylighting (defaultSyntaxMap)
36import System.FilePath (addExtension, replaceExtension, takeExtension)
43import System.FilePath (addExtension, replaceExtension, takeExtension)
37−import Text.Collate.Lang (renderLang)
44+import Text.Pandoc.BCP47 (Lang (..), renderLang)
38−import Text.Pandoc.Builder as B
45+import Text.Pandoc.Builder
39import Text.Pandoc.Class.PandocPure (PandocPure)
46import Text.Pandoc.Class.PandocPure (PandocPure)
40import Text.Pandoc.Class.PandocMonad (PandocMonad (..), getResourcePath,
47import Text.Pandoc.Class.PandocMonad (PandocMonad (..), getResourcePath,
41− readFileFromDirs, report,
48+ readFileFromDirs, report, setResourcePath,
42− setResourcePath)
49+ setTranslations, translateTerm)
43import Text.Pandoc.Error (PandocError (PandocParseError, PandocParsecError))
50import Text.Pandoc.Error (PandocError (PandocParseError, PandocParsecError))
44−import Text.Pandoc.Highlighting (languagesByExtension)
51+import Text.Pandoc.Highlighting (fromListingsLanguage, languagesByExtension)
45import Text.Pandoc.ImageSize (numUnit, showFl)
52import Text.Pandoc.ImageSize (numUnit, showFl)
46import Text.Pandoc.Logging
53import Text.Pandoc.Logging
47import Text.Pandoc.Options
54import Text.Pandoc.Options
48import Text.Pandoc.Parsing hiding (blankline, many, mathDisplay, mathInline,
55import Text.Pandoc.Parsing hiding (blankline, many, mathDisplay, mathInline,
49 optional, space, spaces, withRaw, (<|>))
56 optional, space, spaces, withRaw, (<|>))
50−import Text.Pandoc.Readers.LaTeX.Types (Tok (..), TokType (..))
57+import Text.Pandoc.Readers.LaTeX.Types (ExpansionPoint (..), Macro (..),
58+ ArgSpec (..), Tok (..), TokType (..))
51import Text.Pandoc.Readers.LaTeX.Parsing
59import Text.Pandoc.Readers.LaTeX.Parsing
52−import Text.Pandoc.Readers.LaTeX.Citation (citationCommands, cites)
60+import Text.Pandoc.Readers.LaTeX.Accent (accentCommands)
53−import Text.Pandoc.Readers.LaTeX.Math (dollarsMath, inlineEnvironments,
55− mathDisplay, mathInline,
56− newtheorem, theoremstyle, proof,
58−import Text.Pandoc.Readers.LaTeX.Table (tableEnvironments)
59−import Text.Pandoc.Readers.LaTeX.Macro (macroDef)
60−import Text.Pandoc.Readers.LaTeX.Lang (inlineLanguageCommands,
61+import Text.Pandoc.Readers.LaTeX.Lang (polyglossiaLangToBCP47,
62− babelLangToBCP47, setDefaultLanguage)
62+ babelLangToBCP47)
63−import Text.Pandoc.Readers.LaTeX.SIunitx (siunitxCommands)
63+import Text.Pandoc.Readers.LaTeX.SIunitx
64−import Text.Pandoc.Readers.LaTeX.Inline (acronymCommands, refCommands,
65− nameCommands, charCommands,
67− biblatexInlineCommands,
68− verbCommands, rawInlineOr,
70import Text.Pandoc.Shared
64import Text.Pandoc.Shared
65+import qualified Text.Pandoc.Translations as Translations
71import Text.Pandoc.Walk
66import Text.Pandoc.Walk
72−import Data.List.NonEmpty (nonEmpty)
67+import qualified Text.Pandoc.Builder as B
74-- for debugging:
70-- for debugging:
75-- import Text.Pandoc.Extensions (getDefaultExtensions)
71-- import Text.Pandoc.Extensions (getDefaultExtensions)
@@ -77,17 +73,16 @@ import Data.List.NonEmpty (nonEmpty)
77-- import Debug.Trace (traceShowId)
73-- import Debug.Trace (traceShowId)
79-- | Parse LaTeX from string and return 'Pandoc' document.
75-- | Parse LaTeX from string and return 'Pandoc' document.
80−readLaTeX :: (PandocMonad m, ToSources a)
76+readLaTeX :: PandocMonad m
81 => ReaderOptions -- ^ Reader options
77 => ReaderOptions -- ^ Reader options
82− -> a -- ^ Input to parse
78+ -> Text -- ^ String to parse (assumes @'\n'@ line endings)
83 -> m Pandoc
79 -> m Pandoc
84readLaTeX opts ltx = do
80readLaTeX opts ltx = do
85− let sources = toSources ltx
86 parsed <- runParserT parseLaTeX def{ sOptions = opts } "source"
81 parsed <- runParserT parseLaTeX def{ sOptions = opts } "source"
87− (tokenizeSources sources)
82+ (tokenize "source" (crFilter ltx))
88 case parsed of
83 case parsed of
89 Right result -> return result
84 Right result -> return result
90− Left e -> throwError $ PandocParsecError sources e
85+ Left e -> throwError $ PandocParsecError ltx e
92parseLaTeX :: PandocMonad m => LP m Pandoc
87parseLaTeX :: PandocMonad m => LP m Pandoc
93parseLaTeX = do
88parseLaTeX = do
@@ -98,7 +93,11 @@ parseLaTeX = do
98 let doc' = doc bs
93 let doc' = doc bs
99 let headerLevel (Header n _ _) = [n]
94 let headerLevel (Header n _ _) = [n]
100 headerLevel _ = []
95 headerLevel _ = []
101− let bottomLevel = maybe 1 minimum $ nonEmpty $ query headerLevel doc'
96+#if MIN_VERSION_safe(0,3,18)
97+ let bottomLevel = minimumBound 1 $ query headerLevel doc'
99+ let bottomLevel = minimumDef 1 $ query headerLevel doc'
102 let adjustHeaders m (Header n attr ils) = Header (n+m) attr ils
101 let adjustHeaders m (Header n attr ils) = Header (n+m) attr ils
103 adjustHeaders _ x = x
102 adjustHeaders _ x = x
104 let (Pandoc _ bs') =
103 let (Pandoc _ bs') =
@@ -133,11 +132,11 @@ resolveRefs _ x = x
135rawLaTeXBlock :: (PandocMonad m, HasMacros s, HasReaderOptions s)
134rawLaTeXBlock :: (PandocMonad m, HasMacros s, HasReaderOptions s)
136− => ParserT Sources s m Text
135+ => ParserT Text s m Text
137rawLaTeXBlock = do
136rawLaTeXBlock = do
138 lookAhead (try (char '\\' >> letter))
137 lookAhead (try (char '\\' >> letter))
139 inp <- getInput
138 inp <- getInput
140− let toks = tokenizeSources inp
139+ let toks = tokenize "source" inp
141 snd <$> (rawLaTeXParser toks False (macroDef (const mempty)) blocks
140 snd <$> (rawLaTeXParser toks False (macroDef (const mempty)) blocks
142 <|> rawLaTeXParser toks True
141 <|> rawLaTeXParser toks True
143 (do choice (map controlSeq
142 (do choice (map controlSeq
@@ -164,14 +163,14 @@ beginOrEndCommand = try $ do
164 (txt <> untokenize rawargs)
163 (txt <> untokenize rawargs)
166rawLaTeXInline :: (PandocMonad m, HasMacros s, HasReaderOptions s)
165rawLaTeXInline :: (PandocMonad m, HasMacros s, HasReaderOptions s)
167− => ParserT Sources s m Text
166+ => ParserT Text s m Text
168rawLaTeXInline = do
167rawLaTeXInline = do
169 lookAhead (try (char '\\' >> letter))
168 lookAhead (try (char '\\' >> letter))
170 inp <- getInput
169 inp <- getInput
171− let toks = tokenizeSources inp
170+ let toks = tokenize "source" inp
172 raw <- snd <$>
171 raw <- snd <$>
173 ( rawLaTeXParser toks True
172 ( rawLaTeXParser toks True
174− (mempty <$ (controlSeq "input" >> skipMany rawopt >> braced))
173+ (mempty <$ (controlSeq "input" >> skipMany opt >> braced))
176 <|> rawLaTeXParser toks True (inlineEnvironment <|> inlineCommand')
175 <|> rawLaTeXParser toks True (inlineEnvironment <|> inlineCommand')
@@ -179,11 +178,11 @@ rawLaTeXInline = do
179 finalbraces <- mconcat <$> many (try (string "{}")) -- see #5439
178 finalbraces <- mconcat <$> many (try (string "{}")) -- see #5439
180 return $ raw <> T.pack finalbraces
179 return $ raw <> T.pack finalbraces
182−inlineCommand :: PandocMonad m => ParserT Sources ParserState m Inlines
181+inlineCommand :: PandocMonad m => ParserT Text ParserState m Inlines
183inlineCommand = do
182inlineCommand = do
184 lookAhead (try (char '\\' >> letter))
183 lookAhead (try (char '\\' >> letter))
185 inp <- getInput
184 inp <- getInput
186− let toks = tokenizeSources inp
185+ let toks = tokenize "source" inp
187 fst <$> rawLaTeXParser toks True (inlineEnvironment <|> inlineCommand')
186 fst <$> rawLaTeXParser toks True (inlineEnvironment <|> inlineCommand')
@@ -192,6 +191,12 @@ inlineCommand = do
192word :: PandocMonad m => LP m Inlines
191word :: PandocMonad m => LP m Inlines
193word = str . untoken <$> satisfyTok isWordTok
192word = str . untoken <$> satisfyTok isWordTok
194+regularSymbol :: PandocMonad m => LP m Inlines
195+regularSymbol = str . untoken <$> satisfyTok isRegularSymbol
196+ where isRegularSymbol (Tok _ Symbol t) = not $ T.any isSpecial t
197+ isRegularSymbol _ = False
198+ isSpecial c = c `Set.member` specialChars
195inlineGroup :: PandocMonad m => LP m Inlines
200inlineGroup :: PandocMonad m => LP m Inlines
196inlineGroup = do
201inlineGroup = do
197 ils <- grouped inline
202 ils <- grouped inline
@@ -232,6 +237,16 @@ mkImage options (T.unpack -> src) = do
232 _ -> return src
237 _ -> return src
233 return $ imageWith attr (T.pack src') "" alt
238 return $ imageWith attr (T.pack src') "" alt
240+doxspace :: PandocMonad m => LP m Inlines
242+ (space <$ lookAhead (satisfyTok startsWithLetter)) <|> return mempty
243+ where startsWithLetter (Tok _ Word t) =
245+ Just (c, _) | isLetter c -> True
247+ startsWithLetter _ = False
235removeDoubleQuotes :: Text -> Text
250removeDoubleQuotes :: Text -> Text
236removeDoubleQuotes t =
251removeDoubleQuotes t =
237 Data.Maybe.fromMaybe t $ T.stripPrefix "\"" t >>= T.stripSuffix "\""
252 Data.Maybe.fromMaybe t $ T.stripPrefix "\"" t >>= T.stripSuffix "\""
@@ -281,11 +296,23 @@ quoted' f starter ender = do
281lit :: Text -> LP m Inlines
296lit :: Text -> LP m Inlines
282lit = pure . str
297lit = pure . str
299+enquote :: PandocMonad m => Bool -> Maybe Text -> LP m Inlines
300+enquote starred mblang = do
302+ let lang = mblang >>= babelLangToBCP47
303+ let langspan = case lang of
305+ Just l -> spanWith ("",[],[("lang", renderLang l)])
306+ quoteContext <- sQuoteContext <$> getState
307+ if starred || quoteContext == InDoubleQuote
308+ then singleQuoted . langspan <$> withQuoteContext InSingleQuote tok
309+ else doubleQuoted . langspan <$> withQuoteContext InDoubleQuote tok
284blockquote :: PandocMonad m => Bool -> Maybe Text -> LP m Blocks
311blockquote :: PandocMonad m => Bool -> Maybe Text -> LP m Blocks
285blockquote cvariant mblang = do
312blockquote cvariant mblang = do
286 citepar <- if cvariant
313 citepar <- if cvariant
287 then (\xs -> para (cite xs mempty))
314 then (\xs -> para (cite xs mempty))
288− <$> cites inline NormalCitation False
315+ <$> cites NormalCitation False
289 else option mempty $ para <$> bracketed inline
316 else option mempty $ para <$> bracketed inline
290 let lang = mblang >>= babelLangToBCP47
317 let lang = mblang >>= babelLangToBCP47
291 let langdiv = case lang of
318 let langdiv = case lang of
@@ -296,13 +323,224 @@ blockquote cvariant mblang = do
296 optional $ symbolIn (".:;?!" :: [Char]) -- currently ignored
323 optional $ symbolIn (".:;?!" :: [Char]) -- currently ignored
297 return $ blockQuote . langdiv $ (bs <> citepar)
324 return $ blockQuote . langdiv $ (bs <> citepar)
326+doAcronym :: PandocMonad m => Text -> LP m Inlines
329+ return . mconcat $ [spanWith ("",[],[("acronym-label", untokenize acro),
330+ ("acronym-form", "singular+" <> form)])
331+ $ str $ untokenize acro]
333+doAcronymPlural :: PandocMonad m => Text -> LP m Inlines
334+doAcronymPlural form = do
337+ return . mconcat $ [spanWith ("",[],[("acronym-label", untokenize acro),
338+ ("acronym-form", "plural+" <> form)]) $
339+ mconcat [str $ untokenize acro, plural]]
341+doverb :: PandocMonad m => LP m Inlines
343+ Tok _ Symbol t <- anySymbol
344+ marker <- case T.uncons t of
345+ Just (c, ts) | T.null ts -> return c
348+ code . untokenize <$>
349+ manyTill (notFollowedBy newlineTok >> verbTok marker) (symbol marker)
351+verbTok :: PandocMonad m => Char -> LP m Tok
352+verbTok stopchar = do
353+ t@(Tok pos toktype txt) <- anyTok
354+ case T.findIndex (== stopchar) txt of
357+ let (t1, t2) = T.splitAt i txt
359+ setInput $ Tok (incSourceColumn pos i) Symbol (T.singleton stopchar)
360+ : totoks (incSourceColumn pos (i + 1)) (T.drop 1 t2) ++ inp
361+ return $ Tok pos toktype t1
363+listingsLanguage :: [(Text, Text)] -> Maybe Text
364+listingsLanguage opts =
365+ case lookup "language" opts of
367+ Just l -> fromListingsLanguage l `mplus` Just l
369+dolstinline :: PandocMonad m => LP m Inlines
371+ options <- option [] keyvals
372+ let classes = maybeToList $ listingsLanguage options
373+ doinlinecode classes
375+domintinline :: PandocMonad m => LP m Inlines
378+ cls <- untokenize <$> braced
381+doinlinecode :: PandocMonad m => [Text] -> LP m Inlines
382+doinlinecode classes = do
383+ Tok _ Symbol t <- anySymbol
384+ marker <- case T.uncons t of
385+ Just (c, ts) | T.null ts -> return c
387+ let stopchar = if marker == '{' then '}' else marker
389+ codeWith ("",classes,[]) . T.map nlToSpace . untokenize <$>
390+ manyTill (verbTok stopchar) (symbol stopchar)
392+nlToSpace :: Char -> Char
396+mathDisplay :: Text -> Inlines
397+mathDisplay = displayMath . trimMath
399+mathInline :: Text -> Inlines
400+mathInline = math . trimMath
402+dollarsMath :: PandocMonad m => LP m Inlines
405+ display <- option False (True <$ symbol '$')
406+ (do contents <- try $ untokenize <$> pDollarsMath 0
408+ then mathDisplay contents <$ symbol '$'
409+ else return $ mathInline contents)
410+ <|> (guard display >> return (mathInline ""))
412+-- Int is number of embedded groupings
413+pDollarsMath :: PandocMonad m => Int -> LP m [Tok]
415+ tk@(Tok _ toktype t) <- anyTok
418+ , n == 0 -> return []
421+ (tk :) . (tk' :) <$> pDollarsMath n
422+ | t == "{" -> (tk :) <$> pDollarsMath (n+1)
425+ then (tk :) <$> pDollarsMath (n-1)
427+ _ -> (tk :) <$> pDollarsMath n
431+addPrefix :: [Inline] -> [Citation] -> [Citation]
432+addPrefix p (k:ks) = k {citationPrefix = p ++ citationPrefix k} : ks
435+addSuffix :: [Inline] -> [Citation] -> [Citation]
436+addSuffix s ks@(_:_) =
438+ in init ks ++ [k {citationSuffix = citationSuffix k ++ s}]
441+simpleCiteArgs :: PandocMonad m => LP m [Citation]
442+simpleCiteArgs = try $ do
443+ first <- optionMaybe $ toList <$> opt
444+ second <- optionMaybe $ toList <$> opt
445+ keys <- try $ bgroup *> manyTill citationLabel egroup
446+ let (pre, suf) = case (first , second ) of
447+ (Just s , Nothing) -> (mempty, s )
448+ (Just s , Just t ) -> (s , t )
449+ _ -> (mempty, mempty)
450+ conv k = Citation { citationId = k
451+ , citationPrefix = []
452+ , citationSuffix = []
453+ , citationMode = NormalCitation
455+ , citationNoteNum = 0
457+ return $ addPrefix pre $ addSuffix suf $ map conv keys
459+citationLabel :: PandocMonad m => LP m Text
463+ (many1 (satisfyTok isWordTok <|> symbolIn bibtexKeyChar)
465+ <* optional (symbol ',')
467+ where bibtexKeyChar = ".:;?!`'()/*@_+=-&[]" :: [Char]
469+cites :: PandocMonad m => CitationMode -> Bool -> LP m [Citation]
470+cites mode multi = try $ do
473+ multiprenote <- optionMaybe $ toList <$> paropt
474+ multipostnote <- optionMaybe $ toList <$> paropt
475+ let (pre, suf) = case (multiprenote, multipostnote) of
476+ (Just s , Nothing) -> (mempty, s)
477+ (Nothing , Just t) -> (mempty, t)
478+ (Just s , Just t ) -> (s, t)
479+ _ -> (mempty, mempty)
480+ tempCits <- many1 simpleCiteArgs
482+ (k:ks) -> case ks of
483+ (_:_) -> return $ (addMprenote pre k : init ks) ++
484+ [addMpostnote suf (last ks)]
485+ _ -> return [addMprenote pre (addMpostnote suf k)]
487+ else count 1 simpleCiteArgs
488+ let cs = concat cits
489+ return $ case mode of
490+ AuthorInText -> case cs of
491+ (c:rest) -> c {citationMode = mode} : rest
493+ _ -> map (\a -> a {citationMode = mode}) cs
494+ where mprenote (k:ks) = (k:ks) ++ [Space]
496+ mpostnote (k:ks) = [Str ",", Space] ++ (k:ks)
497+ mpostnote _ = mempty
498+ addMprenote mpn (k:ks) =
499+ let mpnfinal = case citationPrefix k of
500+ (_:_) -> mprenote mpn
502+ in addPrefix mpnfinal (k:ks)
503+ addMprenote _ _ = []
504+ addMpostnote = addSuffix . mpostnote
506+citation :: PandocMonad m => Text -> CitationMode -> Bool -> LP m Inlines
507+citation name mode multi = do
508+ (c,raw) <- withRaw $ cites mode multi
509+ return $ cite c (rawInline "latex" $ "\\" <> name <> untokenize raw)
511+handleCitationPart :: Inlines -> [Citation]
512+handleCitationPart ils =
513+ let isCite Cite{} = True
515+ (pref, rest) = break isCite (toList ils)
517+ (Cite cs _:suff) -> addPrefix pref $ addSuffix suff cs
520+complexNatbibCitation :: PandocMonad m => CitationMode -> LP m Inlines
521+complexNatbibCitation mode = try $ do
523+ withRaw $ concat <$> do
525+ items <- mconcat <$>
526+ many1 (notFollowedBy (symbol ';') >> inline)
529+ return $ map handleCitationPart items
532+ (c:cits) -> return $ cite (c{ citationMode = mode }:cits)
533+ (rawInline "latex" $ "\\citetext" <> untokenize raw)
535+inNote :: Inlines -> Inlines
537+ note $ para $ ils <> str "."
299inlineCommand' :: PandocMonad m => LP m Inlines
539inlineCommand' :: PandocMonad m => LP m Inlines
300inlineCommand' = try $ do
540inlineCommand' = try $ do
301 Tok _ (CtrlSeq name) cmd <- anyControlSeq
541 Tok _ (CtrlSeq name) cmd <- anyControlSeq
302 guard $ name /= "begin" && name /= "end" && name /= "and"
542 guard $ name /= "begin" && name /= "end" && name /= "and"
303− star <- if T.all isAlphaNum name
304− then option "" ("*" <$ symbol '*' <* sp)
543+ star <- option "" ("*" <$ symbol '*' <* sp)
306 overlay <- option "" overlaySpecification
544 overlay <- option "" overlaySpecification
307 let name' = name <> star <> overlay
545 let name' = name <> star <> overlay
308 let names = ordNub [name', name] -- check non-starred as fallback
546 let names = ordNub [name', name] -- check non-starred as fallback
@@ -313,8 +551,28 @@ inlineCommand' = try $ do
313 <|> ignore rawcommand
551 <|> ignore rawcommand
314 lookupListDefault raw names inlineCommands
552 lookupListDefault raw names inlineCommands
316tok :: PandocMonad m => LP m Inlines
555tok :: PandocMonad m => LP m Inlines
317−tok = tokWith inline
556+tok = try $ spaces >> grouped inline <|> inlineCommand' <|> singleChar'
557+ where singleChar' = do
558+ Tok _ _ t <- singleChar
561+opt :: PandocMonad m => LP m Inlines
563+ toks <- try (sp *> bracketedToks <* sp)
564+ -- now parse the toks as inlines
566+ parsed <- runParserT (mconcat <$> many inline) st "bracketed option" toks
568+ Right result -> return result
569+ Left e -> throwError $ PandocParsecError (untokenize toks) e
571+paropt :: PandocMonad m => LP m Inlines
572+paropt = parenWrapped inline
574+inBrackets :: Inlines -> Inlines
575+inBrackets x = str "[" <> x <> str "]"
319unescapeURL :: Text -> Text
577unescapeURL :: Text -> Text
320unescapeURL = T.concat . go . T.splitOn "\\"
578unescapeURL = T.concat . go . T.splitOn "\\"
@@ -327,107 +585,345 @@ unescapeURL = T.concat . go . T.splitOn "\\"
327 , isEscapable c = t
585 , isEscapable c = t
328 | otherwise = "\\" <> t
586 | otherwise = "\\" <> t
588+mathEnvWith :: PandocMonad m
589+ => (Inlines -> a) -> Maybe Text -> Text -> LP m a
590+mathEnvWith f innerEnv name = f . mathDisplay . inner <$> mathEnv name
591+ where inner x = case innerEnv of
593+ Just y -> "\\begin{" <> y <> "}\n" <> x <>
594+ "\\end{" <> y <> "}"
596+mathEnv :: PandocMonad m => Text -> LP m Text
600+ res <- manyTill anyTok (end_ name)
601+ return $ stripTrailingNewlines $ untokenize res
603+inlineEnvironment :: PandocMonad m => LP m Inlines
604+inlineEnvironment = try $ do
606+ name <- untokenize <$> braced
607+ M.findWithDefault mzero name inlineEnvironments
609+inlineEnvironments :: PandocMonad m => M.Map Text (LP m Inlines)
610+inlineEnvironments = M.fromList [
611+ ("displaymath", mathEnvWith id Nothing "displaymath")
612+ , ("math", math <$> mathEnv "math")
613+ , ("equation", mathEnvWith id Nothing "equation")
614+ , ("equation*", mathEnvWith id Nothing "equation*")
615+ , ("gather", mathEnvWith id (Just "gathered") "gather")
616+ , ("gather*", mathEnvWith id (Just "gathered") "gather*")
617+ , ("multline", mathEnvWith id (Just "gathered") "multline")
618+ , ("multline*", mathEnvWith id (Just "gathered") "multline*")
619+ , ("eqnarray", mathEnvWith id (Just "aligned") "eqnarray")
620+ , ("eqnarray*", mathEnvWith id (Just "aligned") "eqnarray*")
621+ , ("align", mathEnvWith id (Just "aligned") "align")
622+ , ("align*", mathEnvWith id (Just "aligned") "align*")
623+ , ("alignat", mathEnvWith id (Just "aligned") "alignat")
624+ , ("alignat*", mathEnvWith id (Just "aligned") "alignat*")
625+ , ("dmath", mathEnvWith id Nothing "dmath")
626+ , ("dmath*", mathEnvWith id Nothing "dmath*")
627+ , ("dgroup", mathEnvWith id (Just "aligned") "dgroup")
628+ , ("dgroup*", mathEnvWith id (Just "aligned") "dgroup*")
629+ , ("darray", mathEnvWith id (Just "aligned") "darray")
630+ , ("darray*", mathEnvWith id (Just "aligned") "darray*")
330inlineCommands :: PandocMonad m => M.Map Text (LP m Inlines)
633inlineCommands :: PandocMonad m => M.Map Text (LP m Inlines)
331−inlineCommands = M.unions
634+inlineCommands =
332− [ accentCommands tok
333− , citationCommands inline
334− , siunitxCommands tok
340− , enquoteCommands tok
341− , inlineLanguageCommands tok
635+ M.union inlineLanguageCommands $
342− , biblatexInlineCommands tok
636+ M.union (accentCommands tok) $
345− rest = M.fromList
637+ M.fromList
346− [ ("emph", extractSpaces emph <$> tok)
638+ [ ("emph", extractSpaces emph <$> tok)
347− , ("textit", extractSpaces emph <$> tok)
639+ , ("textit", extractSpaces emph <$> tok)
348− , ("textsl", extractSpaces emph <$> tok)
640+ , ("textsl", extractSpaces emph <$> tok)
349− , ("textsc", extractSpaces smallcaps <$> tok)
641+ , ("textsc", extractSpaces smallcaps <$> tok)
350− , ("textsf", extractSpaces (spanWith ("",["sans-serif"],[])) <$> tok)
642+ , ("textsf", extractSpaces (spanWith ("",["sans-serif"],[])) <$> tok)
351− , ("textmd", extractSpaces (spanWith ("",["medium"],[])) <$> tok)
643+ , ("textmd", extractSpaces (spanWith ("",["medium"],[])) <$> tok)
352− , ("textrm", extractSpaces (spanWith ("",["roman"],[])) <$> tok)
644+ , ("textrm", extractSpaces (spanWith ("",["roman"],[])) <$> tok)
353− , ("textup", extractSpaces (spanWith ("",["upright"],[])) <$> tok)
645+ , ("textup", extractSpaces (spanWith ("",["upright"],[])) <$> tok)
354− , ("texttt", ttfamily)
646+ , ("texttt", ttfamily)
355− , ("sout", extractSpaces strikeout <$> tok)
647+ , ("sout", extractSpaces strikeout <$> tok)
356− , ("alert", skipopts >> spanWith ("",["alert"],[]) <$> tok) -- beamer
648+ , ("alert", skipopts >> spanWith ("",["alert"],[]) <$> tok) -- beamer
649+ , ("lq", return (str "‘"))
650+ , ("rq", return (str "’"))
651+ , ("textquoteleft", return (str "‘"))
652+ , ("textquoteright", return (str "’"))
653+ , ("textquotedblleft", return (str "“"))
654+ , ("textquotedblright", return (str "”"))
357− , ("textsuperscript", extractSpaces superscript <$> tok)
655+ , ("textsuperscript", extractSpaces superscript <$> tok)
358− , ("textsubscript", extractSpaces subscript <$> tok)
656+ , ("textsubscript", extractSpaces subscript <$> tok)
657+ , ("textbackslash", lit "\\")
658+ , ("backslash", lit "\\")
659+ , ("slash", lit "/")
359− , ("textbf", extractSpaces strong <$> tok)
660+ , ("textbf", extractSpaces strong <$> tok)
360− , ("textnormal", extractSpaces (spanWith ("",["nodecor"],[])) <$> tok)
661+ , ("textnormal", extractSpaces (spanWith ("",["nodecor"],[])) <$> tok)
361− , ("underline", underline <$> tok)
662+ , ("underline", underline <$> tok)
663+ , ("ldots", lit "…")
664+ , ("vdots", lit "\8942")
666+ , ("mdots", lit "…")
669+ , ("label", rawInlineOr "label" dolabel)
670+ , ("ref", rawInlineOr "ref" $ doref "ref")
671+ , ("cref", rawInlineOr "cref" $ doref "ref") -- from cleveref.sty
672+ , ("vref", rawInlineOr "vref" $ doref "ref+page") -- from varioref.sty
673+ , ("eqref", rawInlineOr "eqref" $ doref "eqref") -- from amsmath.sty
362− , ("mbox", rawInlineOr "mbox" $ processHBox <$> tok)
674+ , ("mbox", rawInlineOr "mbox" $ processHBox <$> tok)
363− , ("hbox", rawInlineOr "hbox" $ processHBox <$> tok)
675+ , ("hbox", rawInlineOr "hbox" $ processHBox <$> tok)
364− , ("lettrine", rawInlineOr "lettrine" lettrine)
676+ , ("lettrine", rawInlineOr "lettrine" lettrine)
365− , ("(", mathInline . untokenize <$> manyTill anyTok (controlSeq ")"))
677+ , ("(", mathInline . untokenize <$> manyTill anyTok (controlSeq ")"))
366− , ("[", mathDisplay . untokenize <$> manyTill anyTok (controlSeq "]"))
678+ , ("[", mathDisplay . untokenize <$> manyTill anyTok (controlSeq "]"))
367− , ("ensuremath", mathInline . untokenize <$> braced)
679+ , ("ensuremath", mathInline . untokenize <$> braced)
368− , ("texorpdfstring", const <$> tok <*> tok)
680+ , ("texorpdfstring", const <$> tok <*> tok)
690+ , ("qed", lit "\a0\x25FB")
369− -- old TeX commands
691+ -- old TeX commands
370− , ("em", extractSpaces emph <$> inlines)
692+ , ("em", extractSpaces emph <$> inlines)
371− , ("it", extractSpaces emph <$> inlines)
693+ , ("it", extractSpaces emph <$> inlines)
372− , ("sl", extractSpaces emph <$> inlines)
694+ , ("sl", extractSpaces emph <$> inlines)
373− , ("bf", extractSpaces strong <$> inlines)
695+ , ("bf", extractSpaces strong <$> inlines)
374− , ("tt", code . stringify . toList <$> inlines)
696+ , ("tt", code . stringify . toList <$> inlines)
375− , ("rm", inlines)
697+ , ("rm", inlines)
376− , ("itshape", extractSpaces emph <$> inlines)
698+ , ("itshape", extractSpaces emph <$> inlines)
377− , ("slshape", extractSpaces emph <$> inlines)
699+ , ("slshape", extractSpaces emph <$> inlines)
378− , ("scshape", extractSpaces smallcaps <$> inlines)
700+ , ("scshape", extractSpaces smallcaps <$> inlines)
379− , ("bfseries", extractSpaces strong <$> inlines)
701+ , ("bfseries", extractSpaces strong <$> inlines)
380− , ("MakeUppercase", makeUppercase <$> tok)
702+ , ("MakeUppercase", makeUppercase <$> tok)
381− , ("MakeTextUppercase", makeUppercase <$> tok) -- textcase
703+ , ("MakeTextUppercase", makeUppercase <$> tok) -- textcase
382− , ("uppercase", makeUppercase <$> tok)
704+ , ("uppercase", makeUppercase <$> tok)
383− , ("MakeLowercase", makeLowercase <$> tok)
705+ , ("MakeLowercase", makeLowercase <$> tok)
384− , ("MakeTextLowercase", makeLowercase <$> tok)
706+ , ("MakeTextLowercase", makeLowercase <$> tok)
385− , ("lowercase", makeLowercase <$> tok)
707+ , ("lowercase", makeLowercase <$> tok)
708+ , ("/", pure mempty) -- italic correction
709+ , ("\\", linebreak <$ (do inTableCell <- sInTableCell <$> getState
710+ guard $ not inTableCell
713+ , (",", lit "\8198")
714+ , ("@", pure mempty)
716+ , ("ps", pure $ str "PS." <> space)
717+ , ("TeX", lit "TeX")
718+ , ("LaTeX", lit "LaTeX")
720+ , ("textless", lit "<")
721+ , ("textgreater", lit ">")
386− , ("thanks", skipopts >> note <$> grouped block)
722+ , ("thanks", skipopts >> note <$> grouped block)
387− , ("footnote", skipopts >> note <$> grouped block)
723+ , ("footnote", skipopts >> note <$> grouped block)
388− , ("passthrough", tok) -- \passthrough macro used by latex writer
724+ , ("passthrough", tok) -- \passthrough macro used by latex writer
389− -- for listings
725+ -- for listings
727+ , ("lstinline", dolstinline)
728+ , ("mintinline", domintinline)
730+ , ("url", (\url -> link url "" (str url)) . unescapeURL . untokenize <$>
732+ , ("nolinkurl", code . unescapeURL . untokenize <$> bracedUrl)
733+ , ("href", do url <- bracedUrl
735+ link (unescapeURL $ untokenize url) "" <$> tok)
390− , ("includegraphics", do options <- option [] keyvals
736+ , ("includegraphics", do options <- option [] keyvals
391− src <- braced
737+ src <- braced
392− mkImage options .
738+ mkImage options . unescapeURL . removeDoubleQuotes $
394− removeDoubleQuotes $ untokenize src)
739+ untokenize src)
395− -- hyperref
740+ , ("enquote*", enquote True Nothing)
396− , ("url", (\url -> link url "" (str url)) . unescapeURL . untokenize <$>
741+ , ("enquote", enquote False Nothing)
397− bracedUrl)
742+ -- foreignquote is supposed to use native quote marks
398− , ("nolinkurl", code . unescapeURL . untokenize <$> bracedUrl)
743+ , ("foreignquote*", braced >>= enquote True . Just . untokenize)
399− , ("href", do url <- bracedUrl
744+ , ("foreignquote", braced >>= enquote False . Just . untokenize)
400− sp
745+ -- hypehnquote uses regular quotes
401− link (unescapeURL $ untokenize url) "" <$> tok)
746+ , ("hyphenquote*", braced >>= enquote True . Just . untokenize)
402− , ("hyperlink", hyperlink)
747+ , ("hyphenquote", braced >>= enquote False . Just . untokenize)
403− , ("hyperref", hyperref)
748+ , ("figurename", doTerm Translations.Figure)
749+ , ("prefacename", doTerm Translations.Preface)
750+ , ("refname", doTerm Translations.References)
751+ , ("bibname", doTerm Translations.Bibliography)
752+ , ("chaptername", doTerm Translations.Chapter)
753+ , ("partname", doTerm Translations.Part)
754+ , ("contentsname", doTerm Translations.Contents)
755+ , ("listfigurename", doTerm Translations.ListOfFigures)
756+ , ("listtablename", doTerm Translations.ListOfTables)
757+ , ("indexname", doTerm Translations.Index)
758+ , ("abstractname", doTerm Translations.Abstract)
759+ , ("tablename", doTerm Translations.Table)
760+ , ("enclname", doTerm Translations.Encl)
761+ , ("ccname", doTerm Translations.Cc)
762+ , ("headtoname", doTerm Translations.To)
763+ , ("pagename", doTerm Translations.Page)
764+ , ("seename", doTerm Translations.See)
765+ , ("seealsoname", doTerm Translations.SeeAlso)
766+ , ("proofname", doTerm Translations.Proof)
767+ , ("glossaryname", doTerm Translations.Glossary)
768+ , ("lstlistingname", doTerm Translations.Listing)
769+ , ("cite", citation "cite" NormalCitation False)
770+ , ("Cite", citation "Cite" NormalCitation False)
771+ , ("citep", citation "citep" NormalCitation False)
772+ , ("citep*", citation "citep*" NormalCitation False)
773+ , ("citeal", citation "citeal" NormalCitation False)
774+ , ("citealp", citation "citealp" NormalCitation False)
775+ , ("citealp*", citation "citealp*" NormalCitation False)
776+ , ("autocite", citation "autocite" NormalCitation False)
777+ , ("smartcite", citation "smartcite" NormalCitation False)
778+ , ("footcite", inNote <$> citation "footcite" NormalCitation False)
779+ , ("parencite", citation "parencite" NormalCitation False)
780+ , ("supercite", citation "supercite" NormalCitation False)
781+ , ("footcitetext", inNote <$> citation "footcitetext" NormalCitation False)
782+ , ("citeyearpar", citation "citeyearpar" SuppressAuthor False)
783+ , ("citeyear", citation "citeyear" SuppressAuthor False)
784+ , ("autocite*", citation "autocite*" SuppressAuthor False)
785+ , ("cite*", citation "cite*" SuppressAuthor False)
786+ , ("parencite*", citation "parencite*" SuppressAuthor False)
787+ , ("textcite", citation "textcite" AuthorInText False)
788+ , ("citet", citation "citet" AuthorInText False)
789+ , ("citet*", citation "citet*" AuthorInText False)
790+ , ("citealt", citation "citealt" AuthorInText False)
791+ , ("citealt*", citation "citealt*" AuthorInText False)
792+ , ("textcites", citation "textcites" AuthorInText True)
793+ , ("cites", citation "cites" NormalCitation True)
794+ , ("autocites", citation "autocites" NormalCitation True)
795+ , ("footcites", inNote <$> citation "footcites" NormalCitation True)
796+ , ("parencites", citation "parencites" NormalCitation True)
797+ , ("supercites", citation "supercites" NormalCitation True)
798+ , ("footcitetexts", inNote <$> citation "footcitetexts" NormalCitation True)
799+ , ("Autocite", citation "Autocite" NormalCitation False)
800+ , ("Smartcite", citation "Smartcite" NormalCitation False)
801+ , ("Footcite", inNote <$> citation "Footcite" NormalCitation False)
802+ , ("Parencite", citation "Parencite" NormalCitation False)
803+ , ("Supercite", citation "Supercite" NormalCitation False)
804+ , ("Footcitetext", inNote <$> citation "Footcitetext" NormalCitation False)
805+ , ("Citeyearpar", citation "Citeyearpar" SuppressAuthor False)
806+ , ("Citeyear", citation "Citeyear" SuppressAuthor False)
807+ , ("Autocite*", citation "Autocite*" SuppressAuthor False)
808+ , ("Cite*", citation "Cite*" SuppressAuthor False)
809+ , ("Parencite*", citation "Parencite*" SuppressAuthor False)
810+ , ("Textcite", citation "Textcite" AuthorInText False)
811+ , ("Textcites", citation "Textcites" AuthorInText True)
812+ , ("Cites", citation "Cites" NormalCitation True)
813+ , ("Autocites", citation "Autocites" NormalCitation True)
814+ , ("Footcites", inNote <$> citation "Footcites" NormalCitation True)
815+ , ("Parencites", citation "Parencites" NormalCitation True)
816+ , ("Supercites", citation "Supercites" NormalCitation True)
817+ , ("Footcitetexts", inNote <$> citation "Footcitetexts" NormalCitation True)
818+ , ("citetext", complexNatbibCitation NormalCitation)
819+ , ("citeauthor", (try (tok *> sp *> controlSeq "citetext") *>
820+ complexNatbibCitation AuthorInText)
821+ <|> citation "citeauthor" AuthorInText False)
822+ , ("nocite", mempty <$ (citation "nocite" NormalCitation False >>=
824+ , ("hyperlink", hyperlink)
404− , ("hypertarget", hypertargetInline)
825+ , ("hypertarget", hypertargetInline)
826+ -- glossaries package
827+ , ("gls", doAcronym "short")
828+ , ("Gls", doAcronym "short")
829+ , ("glsdesc", doAcronym "long")
830+ , ("Glsdesc", doAcronym "long")
831+ , ("GLSdesc", doAcronym "long")
832+ , ("acrlong", doAcronym "long")
833+ , ("Acrlong", doAcronym "long")
834+ , ("acrfull", doAcronym "full")
835+ , ("Acrfull", doAcronym "full")
836+ , ("acrshort", doAcronym "abbrv")
837+ , ("Acrshort", doAcronym "abbrv")
838+ , ("glspl", doAcronymPlural "short")
839+ , ("Glspl", doAcronymPlural "short")
840+ , ("glsdescplural", doAcronymPlural "long")
841+ , ("Glsdescplural", doAcronymPlural "long")
842+ , ("GLSdescplural", doAcronymPlural "long")
844+ , ("ac", doAcronym "short")
845+ , ("acf", doAcronym "full")
846+ , ("acs", doAcronym "abbrv")
847+ , ("acl", doAcronym "long")
848+ , ("acp", doAcronymPlural "short")
849+ , ("acfp", doAcronymPlural "full")
850+ , ("acsp", doAcronymPlural "abbrv")
851+ , ("aclp", doAcronymPlural "long")
852+ , ("Ac", doAcronym "short")
853+ , ("Acf", doAcronym "full")
854+ , ("Acs", doAcronym "abbrv")
855+ , ("Acl", doAcronym "long")
856+ , ("Acp", doAcronymPlural "short")
857+ , ("Acfp", doAcronymPlural "full")
858+ , ("Acsp", doAcronymPlural "abbrv")
859+ , ("Aclp", doAcronymPlural "long")
861+ , ("si", skipopts *> dosi tok)
863+ , ("SIrange", doSIrange True tok)
864+ , ("numrange", doSIrange False tok)
865+ , ("numlist", doSInumlist)
405− -- hyphenat
868+ -- hyphenat
869+ , ("bshyp", lit "\\\173")
870+ , ("fshyp", lit "/\173")
871+ , ("dothyp", lit ".\173")
872+ , ("colonhyp", lit ":\173")
406− , ("nohyphens", tok)
874+ , ("nohyphens", tok)
407− , ("textnhtt", ttfamily)
875+ , ("textnhtt", ttfamily)
408− , ("nhttfamily", ttfamily)
876+ , ("nhttfamily", ttfamily)
409− -- LaTeX colors
877+ -- LaTeX colors
410− , ("textcolor", coloredInline "color")
878+ , ("textcolor", coloredInline "color")
411− , ("colorbox", coloredInline "background-color")
879+ , ("colorbox", coloredInline "background-color")
881+ , ("faCheck", lit "\10003")
882+ , ("faClose", lit "\10007")
884+ , ("xspace", doxspace)
412− -- etoolbox
885+ -- etoolbox
413− , ("ifstrequal", ifstrequal)
886+ , ("ifstrequal", ifstrequal)
414− , ("newtoggle", braced >>= newToggle)
887+ , ("newtoggle", braced >>= newToggle)
415− , ("toggletrue", braced >>= setToggle True)
888+ , ("toggletrue", braced >>= setToggle True)
416− , ("togglefalse", braced >>= setToggle False)
889+ , ("togglefalse", braced >>= setToggle False)
417− , ("iftoggle", try $ ifToggle >> inline)
890+ , ("iftoggle", try $ ifToggle >> inline)
892+ , ("RN", romanNumeralUpper)
893+ , ("Rn", romanNumeralLower)
895+ , ("foreignlanguage", foreignlanguage)
418− -- include
896+ -- include
419− , ("input", rawInlineOr "input" $ include "input")
897+ , ("input", rawInlineOr "input" $ include "input")
420− -- soul package
898+ -- soul package
421− , ("ul", underline <$> tok)
899+ , ("ul", underline <$> tok)
422− -- ulem package
900+ -- ulem package
423− , ("uline", underline <$> tok)
901+ , ("uline", underline <$> tok)
424− -- plain tex stuff that should just be passed through as raw tex
902+ -- plain tex stuff that should just be passed through as raw tex
425− , ("ifdim", ifdim)
903+ , ("ifdim", ifdim)
905+ , ("mkbibquote", spanWith nullAttr . doubleQuoted <$> tok)
906+ , ("mkbibemph", spanWith nullAttr . emph <$> tok)
907+ , ("mkbibitalic", spanWith nullAttr . emph <$> tok)
908+ , ("mkbibbold", spanWith nullAttr . strong <$> tok)
910+ spanWith nullAttr . (\x -> str "(" <> x <> str ")") <$> tok)
912+ spanWith nullAttr . (\x -> str "[" <> x <> str "]") <$> tok)
913+ , ("autocap", spanWith nullAttr <$> tok)
914+ , ("textnormal", spanWith ("",["nodecor"],[]) <$> tok)
916+ (\x -> spanWith ("",[],[("bibstring",x)]) (str x)) . untokenize
918+ , ("adddot", pure (str "."))
919+ , ("adddotspace", pure (spanWith nullAttr (str "." <> space)))
920+ , ("addabbrvspace", pure space)
921+ , ("hyphen", pure (str "-"))
428lettrine :: PandocMonad m => LP m Inlines
924lettrine :: PandocMonad m => LP m Inlines
429lettrine = do
925lettrine = do
430− optional rawopt
926+ optional opt
433 return $ extractSpaces (spanWith ("",["lettrine"],[])) x <> smallcaps y
929 return $ extractSpaces (spanWith ("",["lettrine"],[])) x <> smallcaps y
@@ -447,18 +943,32 @@ alterStr :: (Text -> Text) -> Inline -> Inline
447alterStr f (Str xs) = Str (f xs)
943alterStr f (Str xs) = Str (f xs)
448alterStr _ x = x
944alterStr _ x = x
946+foreignlanguage :: PandocMonad m => LP m Inlines
948+ babelLang <- untokenize <$> braced
949+ case babelLangToBCP47 babelLang of
950+ Just lang -> spanWith ("", [], [("lang", renderLang lang)]) <$> tok
953+inlineLanguageCommands :: PandocMonad m => M.Map Text (LP m Inlines)
954+inlineLanguageCommands = M.fromList $ mk <$> M.toList polyglossiaLangToBCP47
956+ mk (polyglossia, bcp47Func) =
957+ ("text" <> polyglossia, inlineLanguage bcp47Func)
959+inlineLanguage :: PandocMonad m => (Text -> Lang) -> LP m Inlines
960+inlineLanguage bcp47Func = do
961+ o <- option "" $ T.filter (\c -> c /= '[' && c /= ']')
963+ let lang = renderLang $ bcp47Func o
964+ extractSpaces (spanWith ("", [], [("lang", lang)])) <$> tok
450hyperlink :: PandocMonad m => LP m Inlines
966hyperlink :: PandocMonad m => LP m Inlines
451hyperlink = try $ do
967hyperlink = try $ do
452 src <- untokenize <$> braced
968 src <- untokenize <$> braced
453 lab <- tok
969 lab <- tok
454 return $ link ("#" <> src) "" lab
970 return $ link ("#" <> src) "" lab
456−hyperref :: PandocMonad m => LP m Inlines
458− url <- (("#" <>) . untokenize <$> try (sp *> bracketedToks <* sp))
459− <|> untokenize <$> (bracedUrl <* bracedUrl <* bracedUrl)
462hypertargetBlock :: PandocMonad m => LP m Blocks
972hypertargetBlock :: PandocMonad m => LP m Blocks
463hypertargetBlock = try $ do
973hypertargetBlock = try $ do
464 ref <- untokenize <$> braced
974 ref <- untokenize <$> braced
@@ -473,6 +983,31 @@ hypertargetInline = try $ do
473 ils <- grouped inline
983 ils <- grouped inline
474 return $ spanWith (ref, [], []) ils
984 return $ spanWith (ref, [], []) ils
986+romanNumeralUpper :: (PandocMonad m) => LP m Inlines
988+ str . toRomanNumeral <$> romanNumeralArg
990+romanNumeralLower :: (PandocMonad m) => LP m Inlines
992+ str . T.toLower . toRomanNumeral <$> romanNumeralArg
994+romanNumeralArg :: (PandocMonad m) => LP m Int
995+romanNumeralArg = spaces *> (parser <|> inBraces)
1005+ s <- untokenize <$> many1 (satisfyTok isWordTok)
1006+ let (digits, rest) = T.span isDigit s
1007+ unless (T.null rest) $
1008+ Prelude.fail "Non-digits in argument to \\Rn or \\RN"
476newToggle :: (Monoid a, PandocMonad m) => [Tok] -> LP m a
1011newToggle :: (Monoid a, PandocMonad m) => [Tok] -> LP m a
477newToggle name = do
1012newToggle name = do
478 updateState $ \st ->
1013 updateState $ \st ->
@@ -503,6 +1038,9 @@ ifToggle = do
503 report $ UndefinedToggle name' pos
1038 report $ UndefinedToggle name' pos
504 return ()
1039 return ()
1041+doTerm :: PandocMonad m => Translations.Term -> LP m Inlines
1042+doTerm term = str <$> translateTerm term
506ifstrequal :: (PandocMonad m, Monoid a) => LP m a
1044ifstrequal :: (PandocMonad m, Monoid a) => LP m a
507ifstrequal = do
1045ifstrequal = do
508 str1 <- tok
1046 str1 <- tok
@@ -523,6 +1061,13 @@ coloredInline stylename = do
523ttfamily :: PandocMonad m => LP m Inlines
1061ttfamily :: PandocMonad m => LP m Inlines
524ttfamily = code . stringify . toList <$> tok
1062ttfamily = code . stringify . toList <$> tok
1064+rawInlineOr :: PandocMonad m => Text -> LP m Inlines -> LP m Inlines
1065+rawInlineOr name' fallback = do
1066+ parseRaw <- extensionEnabled Ext_raw_tex <$> getOption readerExtensions
1068+ then rawInline "latex" <$> getRawCommand name' ("\\" <> name')
526processHBox :: Inlines -> Inlines
1071processHBox :: Inlines -> Inlines
527processHBox = walk convert
1072processHBox = walk convert
@@ -573,67 +1118,67 @@ treatAsInline = Set.fromList
573 , "pagebreak"
1118 , "pagebreak"
1121+label :: PandocMonad m => LP m ()
1125+ updateState $ \st -> st{ sLastLabel = Just $ untokenize t }
1127+dolabel :: PandocMonad m => LP m Inlines
1130+ let refstr = untokenize v
1131+ updateState $ \st ->
1132+ st{ sLastLabel = Just refstr }
1133+ return $ spanWith (refstr,[],[("label", refstr)])
1134+ $ inBrackets $ str $ untokenize v
1136+doref :: PandocMonad m => Text -> LP m Inlines
1139+ let refstr = untokenize v
1140+ return $ linkWith ("",[],[ ("reference-type", cls)
1141+ , ("reference", refstr)])
1144+ (inBrackets $ str refstr)
576lookupListDefault :: (Ord k) => v -> [k] -> M.Map k v -> v
1146lookupListDefault :: (Ord k) => v -> [k] -> M.Map k v -> v
577lookupListDefault d = (fromMaybe d .) . lookupList
1147lookupListDefault d = (fromMaybe d .) . lookupList
578 where lookupList l m = msum $ map (`M.lookup` m) l
1148 where lookupList l m = msum $ map (`M.lookup` m) l
580inline :: PandocMonad m => LP m Inlines
1150inline :: PandocMonad m => LP m Inlines
582− Tok pos toktype t <- lookAhead anyTok
583− let symbolAsString = str . untoken <$> anySymbol
584− let unescapedSymbolAsString =
585− do s <- untoken <$> anySymbol
586− report $ ParsingUnescaped s pos
589− Comment -> mempty <$ comment
1151+inline = (mempty <$ comment)
590− Spaces -> space <$ whitespace
1152+ <|> (space <$ whitespace)
591− Newline -> softbreak <$ endline
1153+ <|> (softbreak <$ endline)
592− Word -> word
1154+ <|> word
593− Esc1 -> str . T.singleton <$> primEscape
1155+ <|> macroDef (rawInline "latex")
594− Esc2 -> str . T.singleton <$> primEscape
1156+ <|> inlineCommand'
595− Symbol ->
1157+ <|> inlineEnvironment
596− case t of
1158+ <|> inlineGroup
597− "-" -> symbol '-' *>
1159+ <|> (symbol '-' *>
598− option (str "-") (symbol '-' *>
1160+ option (str "-") (symbol '-' *>
599− option (str "–") (str "—" <$ symbol '-'))
1161+ option (str "–") (str "—" <$ symbol '-')))
600− "'" -> symbol '\'' *>
1164+ <|> (str "”" <$ try (symbol '\'' >> symbol '\''))
1165+ <|> (str "”" <$ symbol '”')
601− option (str "’") (str "”" <$ symbol '\'')
1166+ <|> (str "’" <$ symbol '\'')
1167+ <|> (str "’" <$ symbol '’')
602− "~" -> str "\160" <$ symbol '~'
1168+ <|> (str "\160" <$ symbol '~')
603− "`" -> doubleQuote <|> singleQuote <|> symbolAsString
604− "\"" -> doubleQuote <|> singleQuote <|> symbolAsString
605− "“" -> doubleQuote <|> symbolAsString
606− "‘" -> singleQuote <|> symbolAsString
607− "$" -> dollarsMath <|> unescapedSymbolAsString
1169+ <|> dollarsMath
608− "|" -> (guardEnabled Ext_literate_haskell *>
1170+ <|> (guardEnabled Ext_literate_haskell *> symbol '|' *> doLHSverb)
609− symbol '|' *> doLHSverb) <|> symbolAsString
1171+ <|> (str . T.singleton <$> primEscape)
610− "{" -> inlineGroup
1172+ <|> regularSymbol
611− "#" -> unescapedSymbolAsString
1173+ <|> (do res <- symbolIn "#^'`\"[]&"
612− "&" -> unescapedSymbolAsString
1174+ pos <- getPosition
613− "_" -> unescapedSymbolAsString
1175+ let s = untoken res
614− "^" -> unescapedSymbolAsString
1176+ report $ ParsingUnescaped s pos
615− "\\" -> mzero
1177+ return $ str s)
618− CtrlSeq _ -> macroDef (rawInline "latex")
620− <|> inlineEnvironment
624inlines :: PandocMonad m => LP m Inlines
1179inlines :: PandocMonad m => LP m Inlines
625inlines = mconcat <$> many inline
1180inlines = mconcat <$> many inline
627−opt :: PandocMonad m => LP m Inlines
629− toks <- try (sp *> bracketedToks <* sp)
630− -- now parse the toks as inlines
632− parsed <- runParserT (mconcat <$> many inline) st "bracketed option" toks
634− Right result -> return result
635− Left e -> throwError $ PandocParsecError (toSources toks) e
637-- block elements:
1182-- block elements:
639preamble :: PandocMonad m => LP m Blocks
1184preamble :: PandocMonad m => LP m Blocks
@@ -727,6 +1272,10 @@ insertIncluded defaultExtension f' = do
727 getInput >>= setInput . (tokenize f contents ++)
1272 getInput >>= setInput . (tokenize f contents ++)
728 updateState dropLatestIncludeFile
1273 updateState dropLatestIncludeFile
1275+addMeta :: PandocMonad m => ToMetaValue a => Text -> a -> LP m ()
1276+addMeta field val = updateState $ \st ->
1277+ st{ sMeta = addMetaField field val $ sMeta st }
730authors :: PandocMonad m => LP m ()
1279authors :: PandocMonad m => LP m ()
731authors = try $ do
1280authors = try $ do
@@ -735,6 +1284,150 @@ authors = try $ do
736 addMeta "author" (map trimInlines auths)
1285 addMeta "author" (map trimInlines auths)
1287+macroDef :: (PandocMonad m, Monoid a) => (Text -> a) -> LP m a
1288+macroDef constructor = do
1289+ (_, s) <- withRaw (commandDef <|> environmentDef)
1290+ (constructor (untokenize s) <$
1291+ guardDisabled Ext_latex_macros)
1293+ where commandDef = do
1294+ (name, macro') <- newcommand <|> letmacro <|> defmacro
1295+ guardDisabled Ext_latex_macros <|>
1296+ updateState (\s -> s{ sMacros = M.insert name macro' (sMacros s) })
1297+ environmentDef = do
1298+ mbenv <- newenvironment
1300+ Nothing -> return ()
1301+ Just (name, macro1, macro2) ->
1302+ guardDisabled Ext_latex_macros <|>
1303+ do updateState $ \s -> s{ sMacros =
1304+ M.insert name macro1 (sMacros s) }
1305+ updateState $ \s -> s{ sMacros =
1306+ M.insert ("end" <> name) macro2 (sMacros s) }
1307+ -- @\newenvironment{envname}[n-args][default]{begin}{end}@
1308+ -- is equivalent to
1309+ -- @\newcommand{\envname}[n-args][default]{begin}@
1310+ -- @\newcommand{\endenvname}@
1312+letmacro :: PandocMonad m => LP m (Text, Macro)
1315+ (name, contents) <- withVerbatimMode $ do
1316+ Tok _ (CtrlSeq name) _ <- anyControlSeq
1317+ optional $ symbol '='
1319+ -- we first parse in verbatim mode, and then expand macros,
1320+ -- because we don't want \let\foo\bar to turn into
1321+ -- \let\foo hello if we have previously \def\bar{hello}
1322+ contents <- bracedOrToken
1323+ return (name, contents)
1324+ contents' <- doMacros' 0 contents
1325+ return (name, Macro ExpandWhenDefined [] Nothing contents')
1327+defmacro :: PandocMonad m => LP m (Text, Macro)
1329+ -- we use withVerbatimMode, because macros are to be expanded
1330+ -- at point of use, not point of definition
1331+ withVerbatimMode $ do
1333+ Tok _ (CtrlSeq name) _ <- anyControlSeq
1334+ argspecs <- many (argspecArg <|> argspecPattern)
1335+ contents <- bracedOrToken
1336+ return (name, Macro ExpandWhenUsed argspecs Nothing contents)
1338+argspecArg :: PandocMonad m => LP m ArgSpec
1340+ Tok _ (Arg i) _ <- satisfyTok isArgTok
1343+argspecPattern :: PandocMonad m => LP m ArgSpec
1345+ Pattern <$> many1 (satisfyTok (\(Tok _ toktype' txt) ->
1346+ (toktype' == Symbol || toktype' == Word) &&
1347+ (txt /= "{" && txt /= "\\" && txt /= "}")))
1349+newcommand :: PandocMonad m => LP m (Text, Macro)
1352+ Tok _ (CtrlSeq mtype) _ <- controlSeq "newcommand" <|>
1353+ controlSeq "renewcommand" <|>
1354+ controlSeq "providecommand" <|>
1355+ controlSeq "DeclareMathOperator" <|>
1356+ controlSeq "DeclareRobustCommand"
1357+ withVerbatimMode $ do
1358+ Tok _ (CtrlSeq name) txt <- do
1359+ optional (symbol '*')
1361+ (symbol '{' *> spaces *> anyControlSeq <* spaces <* symbol '}')
1363+ numargs <- option 0 $ try bracketedNum
1364+ let argspecs = map ArgNum [1..numargs]
1366+ optarg <- option Nothing $ Just <$> try bracketedToks
1368+ contents' <- bracedOrToken
1371+ "DeclareMathOperator" ->
1372+ Tok pos (CtrlSeq "mathop") "\\mathop"
1373+ : Tok pos Symbol "{"
1374+ : Tok pos (CtrlSeq "mathrm") "\\mathrm"
1375+ : Tok pos Symbol "{"
1377+ [ Tok pos Symbol "}", Tok pos Symbol "}" ])
1379+ macros <- sMacros <$> getState
1380+ case M.lookup name macros of
1382+ | mtype == "newcommand" -> do
1383+ report $ MacroAlreadyDefined txt pos
1384+ return (name, macro)
1385+ | mtype == "providecommand" -> return (name, macro)
1386+ _ -> return (name, Macro ExpandWhenUsed argspecs optarg contents)
1388+newenvironment :: PandocMonad m => LP m (Maybe (Text, Macro, Macro))
1391+ Tok _ (CtrlSeq mtype) _ <- controlSeq "newenvironment" <|>
1392+ controlSeq "renewenvironment" <|>
1393+ controlSeq "provideenvironment"
1394+ withVerbatimMode $ do
1395+ optional $ symbol '*'
1397+ name <- untokenize <$> braced
1399+ numargs <- option 0 $ try bracketedNum
1401+ optarg <- option Nothing $ Just <$> try bracketedToks
1402+ let argspecs = map (\i -> ArgNum i) [1..numargs]
1403+ startcontents <- spaces >> bracedOrToken
1404+ endcontents <- spaces >> bracedOrToken
1405+ macros <- sMacros <$> getState
1406+ case M.lookup name macros of
1408+ | mtype == "newenvironment" -> do
1409+ report $ MacroAlreadyDefined name pos
1411+ | mtype == "provideenvironment" ->
1413+ _ -> return $ Just (name,
1414+ Macro ExpandWhenUsed argspecs optarg startcontents,
1415+ Macro ExpandWhenUsed [] Nothing endcontents)
1417+bracketedNum :: PandocMonad m => LP m Int
1419+ ds <- untokenize <$> bracketedToks
1420+ case safeRead ds of
1424+setCaption :: PandocMonad m => LP m ()
1425+setCaption = try $ do
1428+ optional $ try $ spaces *> label
1429+ updateState $ \st -> st{ sCaption = Just ils }
738looseItem :: PandocMonad m => LP m Blocks
1431looseItem :: PandocMonad m => LP m Blocks
739looseItem = do
1432looseItem = do
740 inListItem <- sInListItem <$> getState
1433 inListItem <- sInListItem <$> getState
@@ -748,6 +1441,10 @@ epigraph = do
748 p2 <- grouped block
1441 p2 <- grouped block
749 return $ divWith ("", ["epigraph"], []) (p1 <> p2)
1442 return $ divWith ("", ["epigraph"], []) (p1 <> p2)
1444+resetCaption :: PandocMonad m => LP m ()
1445+resetCaption = updateState $ \st -> st{ sCaption = Nothing
1446+ , sLastLabel = Nothing }
751section :: PandocMonad m => Attr -> Int -> LP m Blocks
1448section :: PandocMonad m => Attr -> Int -> LP m Blocks
752section (ident, classes, kvs) lvl = do
1449section (ident, classes, kvs) lvl = do
753 skipopts
1450 skipopts
@@ -840,7 +1537,7 @@ blockCommands = M.fromList
840 , ("address", mempty <$ (skipopts *> tok >>= addMeta "address"))
1537 , ("address", mempty <$ (skipopts *> tok >>= addMeta "address"))
841 , ("signature", mempty <$ (skipopts *> authors))
1538 , ("signature", mempty <$ (skipopts *> authors))
842 , ("date", mempty <$ (skipopts *> tok >>= addMeta "date"))
1539 , ("date", mempty <$ (skipopts *> tok >>= addMeta "date"))
843− , ("newtheorem", newtheorem inline)
1540+ , ("newtheorem", newtheorem)
844 , ("theoremstyle", theoremstyle)
1541 , ("theoremstyle", theoremstyle)
845 -- KOMA-Script metadata commands
1542 -- KOMA-Script metadata commands
846 , ("extratitle", mempty <$ (skipopts *> tok >>= addMeta "extratitle"))
1543 , ("extratitle", mempty <$ (skipopts *> tok >>= addMeta "extratitle"))
@@ -888,7 +1585,7 @@ blockCommands = M.fromList
888 , ("item", looseItem)
1585 , ("item", looseItem)
889 , ("documentclass", skipopts *> braced *> preamble)
1586 , ("documentclass", skipopts *> braced *> preamble)
890 , ("centerline", para . trimInlines <$> (skipopts *> tok))
1587 , ("centerline", para . trimInlines <$> (skipopts *> tok))
891− , ("caption", mempty <$ setCaption inline)
1588+ , ("caption", mempty <$ setCaption)
892 , ("bibliography", mempty <$ (skipopts *> braced >>=
1589 , ("bibliography", mempty <$ (skipopts *> braced >>=
893 addMeta "bibliography" . splitBibs . untokenize))
1590 addMeta "bibliography" . splitBibs . untokenize))
894 , ("addbibresource", mempty <$ (skipopts *> braced >>=
1591 , ("addbibresource", mempty <$ (skipopts *> braced >>=
@@ -926,8 +1623,7 @@ blockCommands = M.fromList
928environments :: PandocMonad m => M.Map Text (LP m Blocks)
1625environments :: PandocMonad m => M.Map Text (LP m Blocks)
929−environments = M.union (tableEnvironments blocks inline) $
930− M.fromList
1626+environments = M.fromList
931 [ ("document", env "document" blocks <* skipMany anyTok)
1627 [ ("document", env "document" blocks <* skipMany anyTok)
932 , ("abstract", mempty <$ (env "abstract" blocks >>= addMeta "abstract"))
1628 , ("abstract", mempty <$ (env "abstract" blocks >>= addMeta "abstract"))
933 , ("sloppypar", env "sloppypar" blocks)
1629 , ("sloppypar", env "sloppypar" blocks)
@@ -937,6 +1633,13 @@ environments = M.union (tableEnvironments blocks inline) $
937 , ("figure", env "figure" $ skipopts *> figure)
1633 , ("figure", env "figure" $ skipopts *> figure)
938 , ("subfigure", env "subfigure" $ skipopts *> tok *> figure)
1634 , ("subfigure", env "subfigure" $ skipopts *> tok *> figure)
939 , ("center", divWith ("", ["center"], []) <$> env "center" blocks)
1635 , ("center", divWith ("", ["center"], []) <$> env "center" blocks)
1636+ , ("longtable", env "longtable" $
1637+ resetCaption *> simpTable "longtable" False >>= addTableCaption)
1638+ , ("table", env "table" $
1639+ skipopts *> resetCaption *> blocks >>= addTableCaption)
1640+ , ("tabular*", env "tabular*" $ simpTable "tabular*" True)
1641+ , ("tabularx", env "tabularx" $ simpTable "tabularx" True)
1642+ , ("tabular", env "tabular" $ simpTable "tabular" False)
940 , ("quote", blockQuote <$> env "quote" blocks)
1643 , ("quote", blockQuote <$> env "quote" blocks)
941 , ("quotation", blockQuote <$> env "quotation" blocks)
1644 , ("quotation", blockQuote <$> env "quotation" blocks)
942 , ("verse", blockQuote <$> env "verse" blocks)
1645 , ("verse", blockQuote <$> env "verse" blocks)
@@ -959,7 +1662,7 @@ environments = M.union (tableEnvironments blocks inline) $
959 , ("lilypond", rawVerbEnv "lilypond")
1662 , ("lilypond", rawVerbEnv "lilypond")
960 , ("ly", rawVerbEnv "ly")
1663 , ("ly", rawVerbEnv "ly")
961 -- amsthm
1664 -- amsthm
962− , ("proof", proof blocks opt)
1665+ , ("proof", proof)
963 -- etoolbox
1666 -- etoolbox
964 , ("ifstrequal", ifstrequal)
1667 , ("ifstrequal", ifstrequal)
965 , ("newtoggle", braced >>= newToggle)
1668 , ("newtoggle", braced >>= newToggle)
@@ -980,17 +1683,131 @@ filecontents = try $ do
980 st{ sFileContents = M.insert fp txt (sFileContents st) }
1683 st{ sFileContents = M.insert fp txt (sFileContents st) }
981 return mempty
1684 return mempty
1686+theoremstyle :: PandocMonad m => LP m Blocks
1688+ stylename <- untokenize <$> braced
1689+ let mbstyle = case stylename of
1690+ "plain" -> Just PlainStyle
1691+ "definition" -> Just DefinitionStyle
1692+ "remark" -> Just RemarkStyle
1695+ Nothing -> return ()
1696+ Just sty -> updateState $ \s -> s{ sLastTheoremStyle = sty }
1699+newtheorem :: PandocMonad m => LP m Blocks
1701+ number <- option True (False <$ symbol '*' <* sp)
1702+ name <- untokenize <$> braced
1704+ series <- option Nothing $ Just . untokenize <$> bracketedToks
1708+ syncTo <- option Nothing $ Just . untokenize <$> bracketedToks
1709+ sty <- sLastTheoremStyle <$> getState
1710+ let spec = TheoremSpec { theoremName = showName
1711+ , theoremStyle = sty
1712+ , theoremSeries = series
1713+ , theoremSyncTo = syncTo
1714+ , theoremNumber = number
1715+ , theoremLastNum = DottedNum [0] }
1716+ tmap <- sTheoremMap <$> getState
1717+ updateState $ \s -> s{ sTheoremMap =
1718+ M.insert name spec tmap }
1721+proof :: PandocMonad m => LP m Blocks
1723+ title <- option (B.text "Proof") opt
1724+ bs <- env "proof" blocks
1726+ B.divWith ("", ["proof"], []) $
1727+ addQed $ addTitle (B.emph (title <> ".")) bs
1729+addTitle :: Inlines -> Blocks -> Blocks
1731+ case B.toList bs of
1733+ -> B.fromList (Para (B.toList ils ++ (Space : xs)) : rest)
1734+ _ -> B.para ils <> bs
1736+addQed :: Blocks -> Blocks
1738+ case Seq.viewr (B.unMany bs) of
1740+ -> B.Many (s Seq.|> Para (ils ++ B.toList qedSign))
1741+ _ -> bs <> B.para qedSign
1743+ qedSign = B.str "\xa0\x25FB"
983environment :: PandocMonad m => LP m Blocks
1745environment :: PandocMonad m => LP m Blocks
984environment = try $ do
1746environment = try $ do
985 controlSeq "begin"
1747 controlSeq "begin"
986 name <- untokenize <$> braced
1748 name <- untokenize <$> braced
987 M.findWithDefault mzero name environments <|>
1749 M.findWithDefault mzero name environments <|>
988− theoremEnvironment blocks opt name <|>
1750+ theoremEnvironment name <|>
989 if M.member name (inlineEnvironments
1751 if M.member name (inlineEnvironments
990 :: M.Map Text (LP PandocPure Inlines))
1752 :: M.Map Text (LP PandocPure Inlines))
991 then mzero
1753 then mzero
992 else try (rawEnv name) <|> rawVerbEnv name
1754 else try (rawEnv name) <|> rawVerbEnv name
1756+theoremEnvironment :: PandocMonad m => Text -> LP m Blocks
1757+theoremEnvironment name = do
1758+ tmap <- sTheoremMap <$> getState
1759+ case M.lookup name tmap of
1762+ optTitle <- option mempty $ (\x -> space <> "(" <> x <> ")") <$> opt
1763+ mblabel <- option Nothing $ Just . untokenize <$>
1764+ try (spaces >> controlSeq "label" >> spaces >> braced)
1765+ bs <- env name blocks
1767+ if theoremNumber tspec
1769+ let name' = fromMaybe name $ theoremSeries tspec
1770+ num <- getNextNumber
1771+ (maybe (DottedNum [0]) theoremLastNum .
1772+ M.lookup name' . sTheoremMap)
1773+ updateState $ \s ->
1776+ (\spec -> spec{ theoremLastNum = num })
1783+ updateState $ \s ->
1784+ s{ sLabels = M.insert ident
1786+ theoremName tspec <> "\160" <>
1787+ str (renderDottedNum num)) (sLabels s) }
1788+ Nothing -> return ()
1789+ return $ space <> B.text (renderDottedNum num)
1791+ let titleEmph = case theoremStyle tspec of
1792+ PlainStyle -> B.strong
1793+ DefinitionStyle -> B.strong
1794+ RemarkStyle -> B.emph
1795+ let title = titleEmph (theoremName tspec <> number)
1796+ <> optTitle <> "." <> space
1797+ return $ divWith (fromMaybe "" mblabel, [name], []) $ addTitle title
1798+ $ case theoremStyle tspec of
1799+ PlainStyle -> walk italicize bs
1802+italicize :: Block -> Block
1803+italicize x@(Para [Image{}]) = x -- see #6925
1804+italicize (Para ils) = Para [Emph ils]
1805+italicize (Plain ils) = Plain [Emph ils]
1808+env :: PandocMonad m => Text -> LP m a -> LP m a
1809+env name p = p <* end_ name
994rawEnv :: PandocMonad m => Text -> LP m Blocks
1811rawEnv :: PandocMonad m => Text -> LP m Blocks
995rawEnv name = do
1812rawEnv name = do
996 exts <- getOption readerExtensions
1813 exts <- getOption readerExtensions
@@ -1162,8 +1979,7 @@ inputListing = do
1162 let (ident,classes,kvs) = parseListingsOptions options
1979 let (ident,classes,kvs) = parseListingsOptions options
1163 let classes' =
1980 let classes' =
1164 (case listingsLanguage options of
1981 (case listingsLanguage options of
1165− Nothing -> (take 1 (languagesByExtension defaultSyntaxMap
1166− (T.pack $ takeExtension $ T.unpack f)) <>)
1982+ Nothing -> (take 1 (languagesByExtension (T.pack $ takeExtension $ T.unpack f)) <>)
1167 Just _ -> id) classes
1983 Just _ -> id) classes
1168 let firstline = fromMaybe 1 $ lookup "firstline" options >>= safeRead
1984 let firstline = fromMaybe 1 $ lookup "firstline" options >>= safeRead
1169 let lastline = fromMaybe (length codeLines) $
1985 let lastline = fromMaybe (length codeLines) $
@@ -1229,23 +2045,356 @@ orderedList' = try $ do
1229 bs <- listenv "enumerate" (many item)
2045 bs <- listenv "enumerate" (many item)
1230 return $ orderedListWith (start, style, delim) bs
2046 return $ orderedListWith (start, style, delim) bs
2050+hline :: PandocMonad m => LP m ()
2053+ controlSeq "hline" <|>
2055+ controlSeq "toprule" <|>
2056+ controlSeq "bottomrule" <|>
2057+ controlSeq "midrule" <|>
2058+ controlSeq "endhead" <|>
2059+ controlSeq "endfirsthead"
2064+lbreak :: PandocMonad m => LP m Tok
2065+lbreak = (controlSeq "\\" <|> controlSeq "tabularnewline")
2066+ <* skipopts <* spaces
2068+amp :: PandocMonad m => LP m Tok
2071+-- Split a Word into individual Symbols (for parseAligns)
2072+splitWordTok :: PandocMonad m => LP m ()
2076+ (Tok spos Word t : rest) ->
2077+ setInput $ map (Tok spos Symbol . T.singleton) (T.unpack t) <> rest
2080+parseAligns :: PandocMonad m => LP m [(Alignment, ColWidth, ([Tok], [Tok]))]
2081+parseAligns = try $ do
2082+ let maybeBar = skipMany
2083+ (try $ sp *> (() <$ symbol '|' <|> () <$ (symbol '@' >> braced)))
2084+ let cAlign = AlignCenter <$ symbol 'c'
2085+ let lAlign = AlignLeft <$ symbol 'l'
2086+ let rAlign = AlignRight <$ symbol 'r'
2087+ let parAlign = AlignLeft <$ symbol 'p'
2088+ -- aligns from tabularx
2089+ let xAlign = AlignLeft <$ symbol 'X'
2090+ let mAlign = AlignLeft <$ symbol 'm'
2091+ let bAlign = AlignLeft <$ symbol 'b'
2092+ let alignChar = splitWordTok *> ( cAlign <|> lAlign <|> rAlign <|> parAlign
2093+ <|> xAlign <|> mAlign <|> bAlign )
2094+ let alignPrefix = symbol '>' >> braced
2095+ let alignSuffix = symbol '<' >> braced
2096+ let colWidth = try $ do
2098+ ds <- trim . untokenize <$> manyTill anyTok (controlSeq "linewidth")
2101+ return $ safeRead ds
2103+ pref <- option [] alignPrefix
2106+ width <- colWidth <|> option Nothing (do s <- untokenize <$> braced
2108+ report $ SkippedContent s pos
2111+ suff <- option [] alignSuffix
2112+ return (al, width, (pref, suff))
2113+ let starAlign = do -- '*{2}{r}' == 'rr', we just expand like a macro
2116+ ds <- trim . untokenize <$> braced
2119+ case safeRead ds of
2121+ getInput >>= setInput . (mconcat (replicate n spec) ++)
2122+ Nothing -> Prelude.fail $ "Could not parse " <> T.unpack ds <> " as number"
2126+ aligns' <- many $ try $ spaces >> optional starAlign >>
2127+ (alignSpec <* maybeBar)
2131+ return $ map toSpec aligns'
2133+ toColWidth (Just w) | w > 0 = ColWidth w
2134+ toColWidth _ = ColWidthDefault
2135+ toSpec (x, y, z) = (x, toColWidth y, z)
2137+-- N.B. this parser returns a Row that may have erroneous empty cells
2138+-- in it. See the note above fixTableHead for details.
2139+parseTableRow :: PandocMonad m
2140+ => Text -- ^ table environment name
2141+ -> [([Tok], [Tok])] -- ^ pref/suffixes
2143+parseTableRow envname prefsufs = do
2144+ notFollowedBy (spaces *> end_ envname)
2145+ -- add prefixes and suffixes in token stream:
2146+ let celltoks (pref, suff) = do
2147+ prefpos <- getPosition
2148+ contents <- mconcat <$>
2149+ many ( snd <$> withRaw (controlSeq "parbox" >> parbox) -- #5711
2151+ snd <$> withRaw (inlineEnvironment <|> dollarsMath)
2154+ (() <$ amp <|> () <$ lbreak <|> end_ envname)
2157+ suffpos <- getPosition
2158+ option [] (count 1 amp)
2159+ return $ map (setpos prefpos) pref ++ contents ++ map (setpos suffpos) suff
2160+ rawcells <- mapM celltoks prefsufs
2161+ cells <- mapM (parseFromToks parseTableCell) rawcells
2163+ return $ Row nullAttr cells
2165+parseTableCell :: PandocMonad m => LP m Cell
2168+ updateState $ \st -> st{ sInTableCell = True }
2169+ cell' <- multicolumnCell
2171+ <|> parseSimpleCell
2173+ updateState $ \st -> st{ sInTableCell = False }
2177+ -- The parsing of empty cells is important in LaTeX, especially when dealing
2178+ -- with multirow/multicolumn. See #6603.
2179+ parseEmptyCell = spaces $> emptyCell
2181+cellAlignment :: PandocMonad m => LP m Alignment
2182+cellAlignment = skipMany (symbol '|') *> alignment <* skipMany (symbol '|')
2185+ c <- untoken <$> singleChar
2190+ "*" -> AlignDefault
2193+plainify :: Blocks -> Blocks
2194+plainify bs = case toList bs of
2195+ [Para ils] -> plain (fromList ils)
2198+multirowCell :: PandocMonad m => LP m Cell
2199+multirowCell = controlSeq "multirow" >> do
2200+ -- Full prototype for \multirow macro is:
2201+ -- \multirow[vpos]{nrows}[bigstruts]{width}[vmove]{text}
2202+ -- However, everything except `nrows` and `text` make
2203+ -- sense in the context of the Pandoc AST
2204+ _ <- optional $ symbol '[' *> cellAlignment <* symbol ']' -- vertical position
2205+ nrows <- fmap (fromMaybe 1 . safeRead . untokenize) braced
2206+ _ <- optional $ symbol '[' *> manyTill anyTok (symbol ']') -- bigstrut-related
2207+ _ <- symbol '{' *> manyTill anyTok (symbol '}') -- Cell width
2208+ _ <- optional $ symbol '[' *> manyTill anyTok (symbol ']') -- Length used for fine-tuning
2209+ content <- symbol '{' *> (plainify <$> blocks) <* symbol '}'
2210+ return $ cell AlignDefault (RowSpan nrows) (ColSpan 1) content
2212+multicolumnCell :: PandocMonad m => LP m Cell
2213+multicolumnCell = controlSeq "multicolumn" >> do
2214+ span' <- fmap (fromMaybe 1 . safeRead . untokenize) braced
2215+ alignment <- symbol '{' *> cellAlignment <* symbol '}'
2217+ let singleCell = do
2218+ content <- plainify <$> blocks
2219+ return $ cell alignment (RowSpan 1) (ColSpan span') content
2221+ -- Two possible contents: either a \multirow cell, or content.
2222+ -- E.g. \multicol{1}{c}{\multirow{2}{1em}{content}}
2223+ -- Note that a \multirow cell can be nested in a \multicolumn,
2224+ -- but not the other way around. See #6603
2225+ let nestedCell = do
2226+ (Cell _ _ (RowSpan rs) _ bs) <- multirowCell
2233+ symbol '{' *> (nestedCell <|> singleCell) <* symbol '}'
2235+-- Parse a simple cell, i.e. not multirow/multicol
2236+parseSimpleCell :: PandocMonad m => LP m Cell
2237+parseSimpleCell = simpleCell <$> (plainify <$> blocks)
2239+-- LaTeX tables are stored with empty cells underneath multirow cells
2240+-- denoting the grid spaces taken up by them. More specifically, if a
2241+-- cell spans m rows, then it will overwrite all the cells in the
2242+-- columns it spans for (m-1) rows underneath it, requiring padding
2243+-- cells in these places. These padding cells need to be removed for
2244+-- proper table reading. See #6603.
2246+-- These fixTable functions do not otherwise fix up malformed
2247+-- input tables: that is left to the table builder.
2248+fixTableHead :: TableHead -> TableHead
2249+fixTableHead (TableHead attr rows) = TableHead attr rows'
2251+ rows' = fixTableRows rows
2253+fixTableBody :: TableBody -> TableBody
2254+fixTableBody (TableBody attr rhc th tb)
2255+ = TableBody attr rhc th' tb'
2257+ th' = fixTableRows th
2258+ tb' = fixTableRows tb
2260+fixTableRows :: [Row] -> [Row]
2261+fixTableRows = fixTableRows' $ repeat Nothing
2263+ fixTableRows' oldHang (Row attr cells : rs)
2264+ = let (newHang, cells') = fixTableRow oldHang cells
2265+ rs' = fixTableRows' newHang rs
2266+ in Row attr cells' : rs'
2267+ fixTableRows' _ [] = []
2269+-- The overhang is represented as Just (relative cell dimensions) or
2270+-- Nothing for an empty grid space.
2271+fixTableRow :: [Maybe (ColSpan, RowSpan)] -> [Cell] -> ([Maybe (ColSpan, RowSpan)], [Cell])
2272+fixTableRow oldHang cells
2273+ -- If there's overhang, drop cells until their total width meets the
2274+ -- width of the occupied grid spaces (or we run out)
2275+ | (n, prefHang, restHang) <- splitHang oldHang
2277+ = let cells' = dropToWidth getCellW n cells
2278+ (restHang', cells'') = fixTableRow restHang cells'
2279+ in (prefHang restHang', cells'')
2280+ -- Otherwise record the overhang of a pending cell and fix the rest
2282+ | c@(Cell _ _ h w _):cells' <- cells
2285+ oldHang' = dropToWidth getHangW w' oldHang
2286+ (newHang, cells'') = fixTableRow oldHang' cells'
2287+ in (toHang w' h' <> newHang, c : cells'')
2291+ getCellW (Cell _ _ _ w _) = w
2292+ getHangW = maybe 1 fst
2293+ getCS (ColSpan n) = n
2296+ | r > 1 = [Just (c, r)]
2297+ | otherwise = replicate (getCS c) Nothing
2299+ -- Take the prefix of the overhang list representing filled grid
2300+ -- spaces. Also return the remainder and the length of this prefix.
2301+ splitHang = splitHang' 0 id
2303+ splitHang' !n l (Just (c, r):xs)
2304+ = splitHang' (n + c) (l . (toHang c (r-1) ++)) xs
2305+ splitHang' n l xs = (n, l, xs)
2307+ -- Drop list items until the total width of the dropped items
2308+ -- exceeds the passed width.
2309+ dropToWidth _ n l | n < 1 = l
2310+ dropToWidth wproj n (c:cs) = dropToWidth wproj (n - wproj c) cs
2311+ dropToWidth _ _ [] = []
2313+simpTable :: PandocMonad m => Text -> Bool -> LP m Blocks
2314+simpTable envname hasWidthParameter = try $ do
2315+ when hasWidthParameter $ () <$ (spaces >> tok)
2317+ colspecs <- parseAligns
2318+ let (aligns, widths, prefsufs) = unzip3 colspecs
2319+ optional $ controlSeq "caption" *> setCaption
2327+ header' <- option [] . try . fmap (:[]) $
2328+ parseTableRow envname prefsufs <* lbreak <* many1 hline
2330+ rows <- sepEndBy (parseTableRow envname prefsufs)
2331+ (lbreak <* optional (skipMany hline))
2333+ optional $ controlSeq "caption" *> setCaption
2339+ lookAhead $ controlSeq "end" -- make sure we're at end
2340+ let th = fixTableHead $ TableHead nullAttr header'
2341+ let tbs = [fixTableBody $ TableBody nullAttr 0 [] rows]
2342+ let tf = TableFoot nullAttr []
2343+ return $ table emptyCaption (zip aligns widths) th tbs tf
2345+addTableCaption :: PandocMonad m => Blocks -> LP m Blocks
2346+addTableCaption = walkM go
2347+ where go (Table attr c spec th tb tf) = do
2349+ let mblabel = sLastLabel st
2350+ capt <- case (sCaption st, mblabel) of
2351+ (Just ils, Nothing) -> return $ caption Nothing (plain ils)
2352+ (Just ils, Just lab) -> do
2353+ num <- getNextNumber sLastTableNum
2355+ st{ sLastTableNum = num
2356+ , sLabels = M.insert lab
2357+ [Str (renderDottedNum num)]
2359+ return $ caption Nothing (plain ils) -- add number??
2360+ (Nothing, _) -> return c
2361+ let attr' = case (attr, mblabel) of
2362+ ((_,classes,kvs), Just ident) ->
2363+ (ident,classes,kvs)
2365+ return $ addAttrDiv attr' $ Table nullAttr capt spec th tb tf
2368+-- TODO: For now we add a Div to contain table attributes, since
2369+-- most writers don't do anything yet with attributes on Table.
2370+-- This can be removed when that changes.
2371+addAttrDiv :: Attr -> Block -> Block
2372+addAttrDiv ("",[],[]) b = b
2373+addAttrDiv attr b = Div attr [b]
1232block :: PandocMonad m => LP m Blocks
2375block :: PandocMonad m => LP m Blocks
1233block = do
2376block = do
1234− Tok _ toktype _ <- lookAhead anyTok
1235− res <- (case toktype of
1236− Newline -> mempty <$ spaces1
2377+ res <- (mempty <$ spaces1)
1237− Spaces -> mempty <$ spaces1
1238− Comment -> mempty <$ spaces1
1240− CtrlSeq "begin" -> environment
2378+ <|> environment
1241− CtrlSeq _ -> macroDef (rawBlock "latex")
2379+ <|> macroDef (rawBlock "latex")
1242− <|> blockCommand
2380+ <|> blockCommand
1244− <|> paragraph
2381+ <|> paragraph
1245− <|> grouped block
2382+ <|> grouped block
1246 trace (T.take 60 $ tshow $ B.toList res)
2383 trace (T.take 60 $ tshow $ B.toList res)
1247 return res
2384 return res
1249blocks :: PandocMonad m => LP m Blocks
2386blocks :: PandocMonad m => LP m Blocks
1250blocks = mconcat <$> many block
2387blocks = mconcat <$> many block
2389+setDefaultLanguage :: PandocMonad m => LP m Blocks
2390+setDefaultLanguage = do
2391+ o <- option "" $ T.filter (\c -> c /= '[' && c /= ']')
2393+ polylang <- untokenize <$> braced
2394+ case M.lookup polylang polyglossiaLangToBCP47 of
2395+ Nothing -> return mempty -- TODO mzero? warning?
2396+ Just langFunc -> do
2399+ updateState $ setMeta "lang" $ str (renderLang l)