diff --git a/README.markdown b/README.markdown new file mode 100644 --- /dev/null +++ b/README.markdown @@ -0,0 +1,37 @@ +texmath +======= + +texmath is a Haskell library for converting LaTeX math to +MathML and OMML (the math format used in Microsoft Office). +It is used by [pandoc] and [gitit]. + +[pandoc]: http://github.com/jgm/pandoc +[gitit]: http://gitit.net + +You can [try it out online here](http://johnmacfarlane.net/texmath.html). +(Note that the math it produces will be rendered correctly only +if your browser supports MathML. Firefox does; Safari and Chrome do not.) + +By default, only the Haskell library is installed. To install a +test program, `texmath`, use the `test` Cabal flag: + + cabal install -ftest + +`texmath` reads a LaTeX formula from stdin and writes a +standalone xhtml file to stdout. You can run the test suite thus: + + cd tests + sh runtests.sh + +Macro definitions may be included before the formula. + +The `cgi` Cabal flag will cause a CGI binary, `texmath-cgi`, to be +produced: + + cabal install -fcgi + +The file `cgi/texmath.html` contains an example of how it can +be used. + +Thanks to John Lenz for many contributions. + diff --git a/Text/TeXMath/Macros.hs b/Text/TeXMath/Macros.hs --- a/Text/TeXMath/Macros.hs +++ b/Text/TeXMath/Macros.hs @@ -58,7 +58,7 @@ -- | Parses a @\\newcommand@ or @\\renewcommand@ macro definition and -- returns a 'Macro'. pMacroDefinition :: GenParser Char st Macro -pMacroDefinition = newcommand +pMacroDefinition = newcommand <|> declareMathOperator -- | Skip whitespace and comments. pSkipSpaceComments :: GenParser Char st () @@ -103,8 +103,10 @@ newcommand :: GenParser Char st Macro newcommand = try $ do char '\\' - optional $ try $ string "re" - string "newcommand" + -- we ignore differences between these so far: + try (string "newcommand") + <|> try (string "renewcommand") + <|> string "providecommand" pSkipSpaceComments name <- inbraces <|> ctrlseq guard (take 1 name == "\\") @@ -138,7 +140,30 @@ let args' = case opt of Just x -> x : args Nothing -> args - return $ apply args' body + return $ apply args' $ "{" ++ body ++ "}" + +-- | Parser for \DeclareMathOperator(*) command. +declareMathOperator :: GenParser Char st Macro +declareMathOperator = try $ do + string "\\DeclareMathOperator" + pSkipSpaceComments + star <- option "" (string "*") + pSkipSpaceComments + name <- inbraces <|> ctrlseq + guard (take 1 name == "\\") + let name' = drop 1 name + pSkipSpaceComments + body <- inbraces <|> ctrlseq + let defn = "\\DeclareMathOperator" ++ star ++ "{" ++ name ++ "}" ++ + "{" ++ body ++ "}" + return $ Macro defn $ try $ do + char '\\' + string name' + when (all isLetter name') $ + notFollowedBy letter + pSkipSpaceComments + return $ "\\operatorname" ++ star ++ "{" ++ body ++ "}" + apply :: [String] -> String -> String apply args ('#':d:xs) | isDigit d = diff --git a/Text/TeXMath/Pandoc.hs b/Text/TeXMath/Pandoc.hs --- a/Text/TeXMath/Pandoc.hs +++ b/Text/TeXMath/Pandoc.hs @@ -43,7 +43,7 @@ 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] + where addSpace Op x = [x] addSpace Bin x = [medspace, x, medspace] addSpace Rel x = [widespace, x, widespace] addSpace Pun x = [x, thinspace] diff --git a/Text/TeXMath/Parser.hs b/Text/TeXMath/Parser.hs --- a/Text/TeXMath/Parser.hs +++ b/Text/TeXMath/Parser.hs @@ -85,13 +85,44 @@ expr :: TP Exp expr = do optional (try $ symbol "\\displaystyle") - a <- expr1 + (a, convertible) <- try (braces operatorname) -- needed because macros add {} + <|> (expr1 >>= \e -> return (e, False)) + <|> operatorname limits <- limitsIndicator - subSup limits a <|> superOrSubscripted limits a <|> return a + subSup limits convertible a <|> superOrSubscripted limits convertible a <|> return a +-- | Parser for \operatorname command. +-- Returns a tuple of EMathOperator name and Bool depending on the flavor +-- of the command: +-- +-- - True for convertible operator (\operator*) +-- +-- - False otherwise +operatorname :: TP (Exp, Bool) +operatorname = try $ do + symbol "\\operatorname" + convertible <- (char '*' >> spaces >> return True) <|> return False + op <- liftM expToOperatorName texToken + maybe pzero (\s -> return (EMathOperator s, convertible)) op + +-- | Converts identifiers, symbols and numbers to a flat string. +-- Returns Nothing if the expression contains anything else. +expToOperatorName :: Exp -> Maybe String +expToOperatorName e = case e of + EGrouped xs -> liftM concat $ mapM fl xs + _ -> fl e + where fl f = case f of + EIdentifier s -> Just s + -- handle special characters + ESymbol _ "\x2212" -> Just "-" + ESymbol _ "\x02B9" -> Just "'" + ESymbol _ s -> Just s + ENumber s -> Just s + _ -> Nothing + bareSubSup :: TP Exp -bareSubSup = subSup Nothing (EIdentifier "") - <|> superOrSubscripted Nothing (EIdentifier "") +bareSubSup = subSup Nothing False (EIdentifier "") + <|> superOrSubscripted Nothing False (EIdentifier "") limitsIndicator :: TP (Maybe Bool) limitsIndicator = @@ -100,7 +131,7 @@ <|> return Nothing inbraces :: TP Exp -inbraces = liftM EGrouped (braces $ many $ notFollowedBy (char '}') >> expr) +inbraces = liftM EGrouped $ braces $ many $ notFollowedBy (char '}') >> expr texToken :: TP Exp texToken = texSymbol <|> inbraces <|> inbrackets <|> @@ -165,6 +196,7 @@ endLine = try $ do symbol "\\\\" optional inbrackets -- can contain e.g. [1.0in] for a line height, not yet supported + optional $ try $ symbol "\\hline" -- we don't represent the line, but it shouldn't crash parsing return '\n' arrayLine :: TP ArrayLine @@ -271,32 +303,32 @@ isUnderover (EUnder _ (ESymbol Accent "\x23B5")) = True -- \underbracket isUnderover _ = False -subSup :: Maybe Bool -> Exp -> TP Exp -subSup limits a = try $ do +subSup :: Maybe Bool -> Bool -> Exp -> TP Exp +subSup limits convertible a = try $ do let sub1 = symbol "_" >> expr1 let sup1 = symbol "^" >> expr1 (b,c) <- try (do {m <- sub1; n <- sup1; return (m,n)}) <|> (do {n <- sup1; m <- sub1; return (m,n)}) return $ case limits of Just True -> EUnderover a b c - Nothing | isConvertible a -> EDownup a b c + Nothing | convertible || isConvertible a -> EDownup a b c | isUnderover a -> EUnderover a b c _ -> ESubsup a b c -superOrSubscripted :: Maybe Bool -> Exp -> TP Exp -superOrSubscripted limits a = try $ do +superOrSubscripted :: Maybe Bool -> Bool -> Exp -> TP Exp +superOrSubscripted limits convertible a = try $ do c <- oneOf "^_" spaces b <- expr case c of '^' -> return $ case limits of Just True -> EOver a b - Nothing | isConvertible a -> EUp a b + Nothing | convertible || isConvertible a -> EUp a b | isUnderover a -> EOver a b _ -> ESuper a b '_' -> return $ case limits of Just True -> EUnder a b - Nothing | isConvertible a -> EDown a b + Nothing | convertible || isConvertible a -> EDown a b | isUnderover a -> EUnder a b _ -> ESub a b _ -> pzero @@ -533,11 +565,13 @@ ("+", ESymbol Bin "+") , ("-", ESymbol Bin "\x2212") , ("*", ESymbol Bin "*") + , ("@", ESymbol Ord "@") , (",", ESymbol Pun ",") - , (".", ESymbol Pun ".") + , (".", ESymbol Ord ".") , (";", ESymbol Pun ";") - , (":", ESymbol Pun ":") - , ("?", ESymbol Pun "?") + , (":", ESymbol Rel ":") + , ("\\colon", ESymbol Pun ":") + , ("?", ESymbol Ord "?") , (">", ESymbol Rel ">") , ("<", ESymbol Rel "<") , ("!", ESymbol Ord "!") @@ -550,7 +584,7 @@ , ("\\mid", ESymbol Bin "\x2223") , ("\\parallel", ESymbol Rel "\x2225") , ("\\backslash", ESymbol Bin "\x2216") - , ("/", ESymbol Bin "/") + , ("/", ESymbol Ord "/") , ("\\setminus", ESymbol Bin "\\") , ("\\times", ESymbol Bin "\x00D7") , ("\\alpha", EIdentifier "\x03B1") diff --git a/changelog b/changelog new file mode 100644 --- /dev/null +++ b/changelog @@ -0,0 +1,8 @@ +texmath (0.6.6) + + * Insert braces around macro expansions to prevent breakage (#7). + * Support \operatorname and \DeclareMathOperator (rekka) (#17). + * Support \providecommand (#15). + * Fixed spacing bugs in pandoc rendering (#24). + * Ignore \hline at end of array row instead of failing (#19). + diff --git a/tests/macros.omml b/tests/macros.omml --- a/tests/macros.omml +++ b/tests/macros.omml @@ -6,7 +6,11 @@ - 25 + 2 + + + + 5 diff --git a/tests/macros.xhtml b/tests/macros.xhtml --- a/tests/macros.xhtml +++ b/tests/macros.xhtml @@ -6,16 +6,23 @@ - 25 - y - + - 3 - x - + - 3 - x - + - α + 2 + 5 + + y + + + 3 + + + x + + + 3 + + + x + + + α + diff --git a/texmath.cabal b/texmath.cabal --- a/texmath.cabal +++ b/texmath.cabal @@ -1,5 +1,5 @@ Name: texmath -Version: 0.6.5.2 +Version: 0.6.6 Cabal-Version: >= 1.6 Build-type: Custom Synopsis: Conversion of LaTeX math formulas to MathML or OMML. @@ -23,7 +23,9 @@ Author: John MacFarlane Maintainer: jgm@berkeley.edu Homepage: http://github.com/jgm/texmath -Data-Files: cgi/texmath.xhtml, +Extra-source-files: README.markdown, + changelog, + cgi/texmath.xhtml, tests/runtests.sh, tests/01.tex, tests/01.xhtml, tests/01.omml, tests/02.tex, tests/02.xhtml, tests/02.omml,