From c243e5b67b0a3be6a713c6120da24e5e65cb0357 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 25 Apr 2010 12:33:50 -0700 Subject: Use texmath's parser in TexMath module. * This replaces a lot of custom parser code, and expands the tex -> unicode conversion. * The behavior has also changed: if the whole formula can't be converted, the whole formula is left in raw TeX. Previously, pandoc converted parts of the formula to unicode and left other parts in raw TeX. * Added (but not yet exported) readTeXMath', which returns a Maybe. * Updated tests --- src/Text/Pandoc/Readers/TeXMath.hs | 250 ++++++++----------------------------- 1 file changed, 53 insertions(+), 197 deletions(-) (limited to 'src') diff --git a/src/Text/Pandoc/Readers/TeXMath.hs b/src/Text/Pandoc/Readers/TeXMath.hs index 080354be1..40cf39987 100644 --- a/src/Text/Pandoc/Readers/TeXMath.hs +++ b/src/Text/Pandoc/Readers/TeXMath.hs @@ -28,208 +28,64 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA Conversion of TeX math to a list of 'Pandoc' inline elements. -} module Text.Pandoc.Readers.TeXMath ( - readTeXMath + readTeXMath ) where import Text.ParserCombinators.Parsec import Text.Pandoc.Definition +import Text.TeXMath.Parser --- | Converts a string of raw TeX math to a list of 'Pandoc' inlines. +-- | Converts a raw TeX math formula to a list of 'Pandoc' inlines. +-- Defaults to raw formula between @$@ characters if entire formula +-- can't be converted. readTeXMath :: String -- ^ String to parse (assumes @'\n'@ line endings) -> [Inline] -readTeXMath inp = case parse teXMath ("formula: " ++ inp) inp of - Left _ -> [Str inp] -- if unparseable, just include original - Right res -> res - -teXMath :: GenParser Char st [Inline] -teXMath = manyTill mathPart eof >>= return . concat - -mathPart :: GenParser Char st [Inline] -mathPart = whitespace <|> superscript <|> subscript <|> symbol <|> - argument <|> digits <|> letters <|> misc - -whitespace :: GenParser Char st [Inline] -whitespace = many1 space >> return [] - -symbol :: GenParser Char st [Inline] -symbol = try $ do - char '\\' - res <- many1 letter - case lookup res teXsymbols of - Just m -> return [Str m] - Nothing -> return [Str $ "\\" ++ res] - -argument :: GenParser Char st [Inline] -argument = try $ do - char '{' - res <- many mathPart - char '}' - return $ if null res - then [Str " "] - else [Str "{"] ++ concat res ++ [Str "}"] - -digits :: GenParser Char st [Inline] -digits = do - res <- many1 digit - return [Str res] - -letters :: GenParser Char st [Inline] -letters = do - res <- many1 letter - return [Emph [Str res]] - -misc :: GenParser Char st [Inline] -misc = do - res <- noneOf "}" - return [Str [res]] - -scriptArg :: GenParser Char st [Inline] -scriptArg = try $ do - (try (do{char '{'; r <- many mathPart; char '}'; return $ concat r})) - <|> symbol - <|> (do{c <- (letter <|> digit); return [Str [c]]}) - -superscript :: GenParser Char st [Inline] -superscript = try $ do - char '^' - arg <- scriptArg - return [Superscript arg] - -subscript :: GenParser Char st [Inline] -subscript = try $ do - char '_' - arg <- scriptArg - return [Subscript arg] - -withThinSpace :: String -> String -withThinSpace str = "\x2009" ++ str ++ "\x2009" - -teXsymbols :: [(String, String)] -teXsymbols = - [("alpha","\x3B1") - ,("beta", "\x3B2") - ,("chi", "\x3C7") - ,("delta", "\x3B4") - ,("Delta", "\x394") - ,("epsilon", "\x3B5") - ,("varepsilon", "\x25B") - ,("eta", "\x3B7") - ,("gamma", "\x3B3") - ,("Gamma", "\x393") - ,("iota", "\x3B9") - ,("kappa", "\x3BA") - ,("lambda", "\x3BB") - ,("Lambda", "\x39B") - ,("mu", "\x3BC") - ,("nu", "\x3BD") - ,("omega", "\x3C9") - ,("Omega", "\x3A9") - ,("phi", "\x3C6") - ,("varphi", "\x3D5") - ,("Phi", "\x3A6") - ,("pi", "\x3C0") - ,("Pi", "\x3A0") - ,("psi", "\x3C8") - ,("Psi", "\x3A8") - ,("rho", "\x3C1") - ,("sigma", "\x3C3") - ,("Sigma", "\x3A3") - ,("tau", "\x3C4") - ,("theta", "\x3B8") - ,("vartheta", "\x3D1") - ,("Theta", "\x398") - ,("upsilon", "\x3C5") - ,("xi", "\x3BE") - ,("Xi", "\x39E") - ,("zeta", "\x3B6") - ,("ne", "\x2260") - ,("lt", withThinSpace "<") - ,("le", withThinSpace "\x2264") - ,("leq", withThinSpace "\x2264") - ,("ge", withThinSpace "\x2265") - ,("geq", withThinSpace "\x2265") - ,("prec", withThinSpace "\x227A") - ,("succ", withThinSpace "\x227B") - ,("preceq", withThinSpace "\x2AAF") - ,("succeq", withThinSpace "\x2AB0") - ,("in", withThinSpace "\x2208") - ,("notin", withThinSpace "\x2209") - ,("subset", withThinSpace "\x2282") - ,("supset", withThinSpace "\x2283") - ,("subseteq", withThinSpace "\x2286") - ,("supseteq", withThinSpace "\x2287") - ,("equiv", withThinSpace "\x2261") - ,("cong", withThinSpace "\x2245") - ,("approx", withThinSpace "\x2248") - ,("propto", withThinSpace "\x221D") - ,("cdot", withThinSpace "\x22C5") - ,("star", withThinSpace "\x22C6") - ,("backslash", "\\") - ,("times", withThinSpace "\x00D7") - ,("divide", withThinSpace "\x00F7") - ,("circ", withThinSpace "\x2218") - ,("oplus", withThinSpace "\x2295") - ,("otimes", withThinSpace "\x2297") - ,("odot", withThinSpace "\x2299") - ,("sum", "\x2211") - ,("prod", "\x220F") - ,("wedge", withThinSpace "\x2227") - ,("bigwedge", withThinSpace "\x22C0") - ,("vee", withThinSpace "\x2228") - ,("bigvee", withThinSpace "\x22C1") - ,("cap", withThinSpace "\x2229") - ,("bigcap", withThinSpace "\x22C2") - ,("cup", withThinSpace "\x222A") - ,("bigcup", withThinSpace "\x22C3") - ,("neg", "\x00AC") - ,("implies", withThinSpace "\x21D2") - ,("iff", withThinSpace "\x21D4") - ,("forall", "\x2200") - ,("exists", "\x2203") - ,("bot", "\x22A5") - ,("top", "\x22A4") - ,("vdash", "\x22A2") - ,("models", withThinSpace "\x22A8") - ,("uparrow", "\x2191") - ,("downarrow", "\x2193") - ,("rightarrow", withThinSpace "\x2192") - ,("to", withThinSpace "\x2192") - ,("rightarrowtail", "\x21A3") - ,("twoheadrightarrow", withThinSpace "\x21A0") - ,("twoheadrightarrowtail", withThinSpace "\x2916") - ,("mapsto", withThinSpace "\x21A6") - ,("leftarrow", withThinSpace "\x2190") - ,("leftrightarrow", withThinSpace "\x2194") - ,("Rightarrow", withThinSpace "\x21D2") - ,("Leftarrow", withThinSpace "\x21D0") - ,("Leftrightarrow", withThinSpace "\x21D4") - ,("partial", "\x2202") - ,("nabla", "\x2207") - ,("pm", "\x00B1") - ,("emptyset", "\x2205") - ,("infty", "\x221E") - ,("aleph", "\x2135") - ,("ldots", "...") - ,("therefore", "\x2234") - ,("angle", "\x2220") - ,("quad", "\x00A0\x00A0") - ,("cdots", "\x22EF") - ,("vdots", "\x22EE") - ,("ddots", "\x22F1") - ,("diamond", "\x22C4") - ,("Box", "\x25A1") - ,("lfloor", "\x230A") - ,("rfloor", "\x230B") - ,("lceiling", "\x2308") - ,("rceiling", "\x2309") - ,("langle", "\x2329") - ,("rangle", "\x232A") - ,("int", "\8747") - ,("{", "{") - ,("}", "}") - ,("[", "[") - ,("]", "]") - ,("|", "|") - ,("||", "||") - ] +readTeXMath inp = case readTeXMath' inp of + Nothing -> [Str ("$" ++ inp ++ "$")] + Just res -> res + +-- | Like 'readTeXMath', but without the default. +readTeXMath' :: String -- ^ String to parse (assumes @'\n'@ line endings) + -> Maybe [Inline] +readTeXMath' inp = case parse formula "formula" inp of + Left _ -> Just [Str inp] + Right exps -> expsToInlines exps + +expsToInlines :: [Exp] -> Maybe [Inline] +expsToInlines xs = do + res <- mapM expToInlines xs + return (concat res) + +expToInlines :: Exp -> Maybe [Inline] +expToInlines (ENumber s) = Just [Str s] +expToInlines (EIdentifier s) = Just [Emph [Str s]] +expToInlines (EMathOperator s) = Just [Str s] +expToInlines (ESymbol t s) = Just $ addSpace t (Str s) + where addSpace Op x = [x, thinspace] + addSpace Bin x = [medspace, x, medspace] + addSpace Rel x = [widespace, x, widespace] + addSpace Pun x = [x, thinspace] + addSpace _ x = [x] + thinspace = Str "\x2006" + medspace = Str "\x2005" + widespace = Str "\x2004" +expToInlines (EStretchy x) = expToInlines x +expToInlines (EGrouped xs) = expsToInlines xs +expToInlines (ESpace _) = Just [Str " "] -- variable widths not supported +expToInlines (EBinary _ _ _) = Nothing +expToInlines (ESub x y) = do + x' <- expToInlines x + y' <- expToInlines y + return $ x' ++ [Subscript y'] +expToInlines (ESuper x y) = do + x' <- expToInlines x + y' <- expToInlines y + return $ x' ++ [Superscript y'] +expToInlines (ESubsup x y z) = do + x' <- expToInlines x + y' <- expToInlines y + z' <- expToInlines z + return $ x' ++ [Subscript y'] ++ [Superscript z'] +expToInlines (EText _ x) = Just [Emph [Str x]] +expToInlines _ = Nothing -- cgit v1.2.3