diff options
Diffstat (limited to 'src/Text/Pandoc/Writers')
| -rw-r--r-- | src/Text/Pandoc/Writers/ConTeXt.hs | 8 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Docbook.hs | 5 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Docx.hs | 89 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/EPUB.hs | 745 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/FB2.hs | 6 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/HTML.hs | 59 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/LaTeX.hs | 99 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Markdown.hs | 38 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/MediaWiki.hs | 2 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/ODT.hs | 53 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/OpenDocument.hs | 26 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Org.hs | 2 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/RST.hs | 2 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Shared.hs | 30 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Texinfo.hs | 6 | ||||
| -rw-r--r-- | src/Text/Pandoc/Writers/Textile.hs | 2 |
16 files changed, 935 insertions, 237 deletions
diff --git a/src/Text/Pandoc/Writers/ConTeXt.hs b/src/Text/Pandoc/Writers/ConTeXt.hs index 0379f8b0a..3095cf508 100644 --- a/src/Text/Pandoc/Writers/ConTeXt.hs +++ b/src/Text/Pandoc/Writers/ConTeXt.hs @@ -130,7 +130,7 @@ blockToConTeXt (Plain lst) = inlineListToConTeXt lst -- title beginning with fig: indicates that the image is a figure blockToConTeXt (Para [Image txt (src,'f':'i':'g':':':_)]) = do capt <- inlineListToConTeXt txt - return $ blankline $$ "\\placefigure[here,nonumber]" <> braces capt <> + return $ blankline $$ "\\placefigure" <> braces capt <> braces ("\\externalfigure" <> brackets (text src)) <> blankline blockToConTeXt (Para lst) = do contents <- inlineListToConTeXt lst @@ -205,9 +205,9 @@ blockToConTeXt (Table caption aligns widths heads rows) = do else liftM ($$ "\\HL") $ tableRowToConTeXt heads captionText <- inlineListToConTeXt caption rows' <- mapM tableRowToConTeXt rows - return $ "\\placetable" <> brackets ("here" <> if null caption - then ",none" - else "") + return $ "\\placetable" <> (if null caption + then brackets "none" + else empty) <> braces captionText $$ "\\starttable" <> brackets (text colDescriptors) $$ "\\HL" $$ headers $$ diff --git a/src/Text/Pandoc/Writers/Docbook.hs b/src/Text/Pandoc/Writers/Docbook.hs index dad83d7bb..02d875be3 100644 --- a/src/Text/Pandoc/Writers/Docbook.hs +++ b/src/Text/Pandoc/Writers/Docbook.hs @@ -85,8 +85,9 @@ writeDocbook opts (Pandoc meta blocks) = auths' = map (authorToDocbook opts) $ docAuthors meta meta' = B.setMeta "author" auths' meta Just metadata = metaToJSON opts - (Just . render colwidth . blocksToDocbook opts) - (Just . render colwidth . inlinesToDocbook opts) + (Just . render colwidth . (vcat . + (map (elementToDocbook opts' startLvl)) . hierarchicalize)) + (Just . render colwidth . inlinesToDocbook opts') meta' main = render' $ vcat (map (elementToDocbook opts' startLvl) elements) context = defField "body" main diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 0fdea0a7a..2a834c2da 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -29,7 +29,8 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA Conversion of 'Pandoc' documents to docx. -} module Text.Pandoc.Writers.Docx ( writeDocx ) where -import Data.List ( intercalate, groupBy ) +import Data.Maybe (fromMaybe) +import Data.List ( intercalate, isPrefixOf, isSuffixOf ) import qualified Data.ByteString as B import qualified Data.ByteString.Lazy as BL import qualified Data.ByteString.Lazy.Char8 as BL8 @@ -42,6 +43,7 @@ import Text.Pandoc.Definition import Text.Pandoc.Generic import Text.Pandoc.ImageSize import Text.Pandoc.Shared hiding (Element) +import Text.Pandoc.Writers.Shared (fixDisplayMath) import Text.Pandoc.Options import Text.Pandoc.Readers.TeXMath import Text.Pandoc.Highlighting ( highlight ) @@ -55,8 +57,8 @@ import Data.Unique (hashUnique, newUnique) import System.Random (randomRIO) import Text.Printf (printf) import qualified Control.Exception as E -import System.FilePath (takeExtension) -import Text.Pandoc.MIME (getMimeType) +import Text.Pandoc.MIME (getMimeType, extensionFromMimeType) +import Control.Applicative ((<|>)) data WriterState = WriterState{ stTextProperties :: [Element] @@ -130,7 +132,8 @@ writeDocx opts doc@(Pandoc meta _) = do let mkOverrideNode (part', contentType') = mknode "Override" [("PartName",part'),("ContentType",contentType')] () let mkImageOverride (_, imgpath, mbMimeType, _, _) = - mkOverrideNode ("/word/" ++ imgpath, maybe "application/octet-stream" id mbMimeType) + mkOverrideNode ("/word/" ++ imgpath, + fromMaybe "application/octet-stream" mbMimeType) let overrides = map mkOverrideNode [("/word/webSettings.xml", "application/vnd.openxmlformats-officedocument.wordprocessingml.webSettings+xml") @@ -231,10 +234,11 @@ writeDocx opts doc@(Pandoc meta _) = do ,("xmlns:dcmitype","http://purl.org/dc/dcmitype/") ,("xmlns:xsi","http://www.w3.org/2001/XMLSchema-instance")] $ mknode "dc:title" [] (stringify $ docTitle meta) - : mknode "dcterms:created" [("xsi:type","dcterms:W3CDTF")] - (maybe "" id $ normalizeDate $ stringify $ docDate meta) - : mknode "dcterms:modified" [("xsi:type","dcterms:W3CDTF")] () -- put current time here - : map (mknode "dc:creator" [] . stringify) (docAuthors meta) + : mknode "dc:creator" [] (intercalate "; " (map stringify $ docAuthors meta)) + : maybe [] + (\x -> [ mknode "dcterms:created" [("xsi:type","dcterms:W3CDTF")] $ x + , mknode "dcterms:modified" [("xsi:type","dcterms:W3CDTF")] $ x + ]) (normalizeDate $ stringify $ docDate meta) let docPropsEntry = toEntry docPropsPath epochtime $ renderXml docProps let relsPath = "_rels/.rels" @@ -247,7 +251,7 @@ writeDocx opts doc@(Pandoc meta _) = do ,("Type","http://schemas.openxmlformats.org/officeDocument/2006/relationships/extended-properties") ,("Target","docProps/app.xml")] , [("Id","rId3") - ,("Type","http://schemas.openxmlformats.org/officedocument/2006/relationships/metadata/core-properties") + ,("Type","http://schemas.openxmlformats.org/package/2006/relationships/metadata/core-properties") ,("Target","docProps/core.xml")] ] let relsEntry = toEntry relsPath epochtime $ renderXml rels @@ -259,6 +263,12 @@ writeDocx opts doc@(Pandoc meta _) = do fontTableEntry <- entryFromArchive "word/fontTable.xml" settingsEntry <- entryFromArchive "word/settings.xml" webSettingsEntry <- entryFromArchive "word/webSettings.xml" + let miscRels = [ f | f <- filesInArchive refArchive + , "word/_rels/" `isPrefixOf` f + , ".xml.rels" `isSuffixOf` f + , f /= "word/_rels/document.xml.rels" + , f /= "word/_rels/footnotes.xml.rels" ] + miscRelEntries <- mapM entryFromArchive miscRels -- Create archive let archive = foldr addEntryToArchive emptyArchive $ @@ -266,7 +276,7 @@ writeDocx opts doc@(Pandoc meta _) = do footnoteRelEntry : numEntry : styleEntry : footnotesEntry : docPropsEntry : docPropsAppEntry : themeEntry : fontTableEntry : settingsEntry : webSettingsEntry : - imageEntries + imageEntries ++ miscRelEntries return $ fromArchive archive styleToOpenXml :: Style -> [Element] @@ -321,7 +331,7 @@ mkNum markers marker numid = NumberMarker _ _ start -> map (\lvl -> mknode "w:lvlOverride" [("w:ilvl",show (lvl :: Int))] $ mknode "w:startOverride" [("w:val",show start)] ()) [0..6] - where absnumid = maybe 0 id $ M.lookup marker markers + where absnumid = fromMaybe 0 $ M.lookup marker markers mkAbstractNum :: (ListMarker,Int) -> IO Element mkAbstractNum (marker,numid) = do @@ -637,7 +647,12 @@ formattedString str = do inlineToOpenXML :: WriterOptions -> Inline -> WS [Element] inlineToOpenXML _ (Str str) = formattedString str inlineToOpenXML opts Space = inlineToOpenXML opts (Str " ") -inlineToOpenXML opts (Span _ ils) = inlinesToOpenXML opts ils +inlineToOpenXML opts (Span (_,classes,_) ils) = do + let off x = withTextProp (mknode x [("w:val","0")] ()) + ((if "csl-no-emph" `elem` classes then off "w:i" else id) . + (if "csl-no-strong" `elem` classes then off "w:b" else id) . + (if "csl-no-smallcaps" `elem` classes then off "w:smallCaps" else id)) + $ inlinesToOpenXML opts ils inlineToOpenXML opts (Strong lst) = withTextProp (mknode "w:b" [] ()) $ inlinesToOpenXML opts lst inlineToOpenXML opts (Emph lst) = @@ -736,7 +751,7 @@ inlineToOpenXML opts (Image alt (src, tit)) = do liftIO $ warn $ "Could not find image `" ++ src ++ "', skipping..." -- emit alt text inlinesToOpenXML opts alt - Right (img, _) -> do + Right (img, mt) -> do ident <- ("rId"++) `fmap` getUniqueId let size = imageSize img let (xpt,ypt) = maybe (120,120) sizeInPoints size @@ -775,19 +790,21 @@ inlineToOpenXML opts (Image alt (src, tit)) = do , mknode "wp:effectExtent" [("b","0"),("l","0"),("r","0"),("t","0")] () , mknode "wp:docPr" [("descr",tit),("id","1"),("name","Picture")] () , graphic ] - let imgext = case imageType img of - Just Png -> ".png" - Just Jpeg -> ".jpeg" - Just Gif -> ".gif" - Just Pdf -> ".pdf" - Just Eps -> ".eps" - Nothing -> takeExtension src + let imgext = case mt >>= extensionFromMimeType of + Just x -> '.':x + Nothing -> case imageType img of + Just Png -> ".png" + Just Jpeg -> ".jpeg" + Just Gif -> ".gif" + Just Pdf -> ".pdf" + Just Eps -> ".eps" + Nothing -> "" if null imgext then -- without an extension there is no rule for content type inlinesToOpenXML opts alt -- return alt to avoid corrupted docx else do let imgpath = "media/" ++ ident ++ imgext - let mbMimeType = getMimeType imgpath + let mbMimeType = mt <|> getMimeType imgpath -- insert mime type to use in constructing [Content_Types].xml modify $ \st -> st{ stImages = M.insert src (ident, imgpath, mbMimeType, imgElt, img) @@ -799,30 +816,8 @@ br = mknode "w:r" [] [mknode "w:br" [("w:type","textWrapping")] () ] parseXml :: Archive -> String -> IO Element parseXml refArchive relpath = - case (findEntryByPath relpath refArchive >>= parseXMLDoc . UTF8.toStringLazy . fromEntry) of - Just d -> return d + case findEntryByPath relpath refArchive of + Just e -> case parseXMLDoc $ UTF8.toStringLazy $ fromEntry e of + Just d -> return d + Nothing -> fail $ relpath ++ " corrupt in reference docx" Nothing -> fail $ relpath ++ " missing in reference docx" - -isDisplayMath :: Inline -> Bool -isDisplayMath (Math DisplayMath _) = True -isDisplayMath _ = False - -stripLeadingTrailingSpace :: [Inline] -> [Inline] -stripLeadingTrailingSpace = go . reverse . go . reverse - where go (Space:xs) = xs - go xs = xs - -fixDisplayMath :: Block -> Block -fixDisplayMath (Plain lst) - | any isDisplayMath lst && not (all isDisplayMath lst) = - -- chop into several paragraphs so each displaymath is its own - Div ("",["math"],[]) $ map (Plain . stripLeadingTrailingSpace) $ - groupBy (\x y -> (isDisplayMath x && isDisplayMath y) || - not (isDisplayMath x || isDisplayMath y)) lst -fixDisplayMath (Para lst) - | any isDisplayMath lst && not (all isDisplayMath lst) = - -- chop into several paragraphs so each displaymath is its own - Div ("",["math"],[]) $ map (Para . stripLeadingTrailingSpace) $ - groupBy (\x y -> (isDisplayMath x && isDisplayMath y) || - not (isDisplayMath x || isDisplayMath y)) lst -fixDisplayMath x = x diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index ac0e7610c..a48300939 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -1,4 +1,4 @@ -{-# LANGUAGE PatternGuards, CPP #-} +{-# LANGUAGE PatternGuards, CPP, ScopedTypeVariables #-} {- Copyright (C) 2010 John MacFarlane <jgm@berkeley.edu> @@ -30,16 +30,18 @@ Conversion of 'Pandoc' documents to EPUB. -} module Text.Pandoc.Writers.EPUB ( writeEPUB ) where import Data.IORef -import Data.Maybe ( fromMaybe, isNothing ) +import qualified Data.Map as M +import Data.Maybe ( fromMaybe ) import Data.List ( isInfixOf, intercalate ) import System.Environment ( getEnv ) import Text.Printf (printf) import System.FilePath ( (</>), takeBaseName, takeExtension, takeFileName ) import qualified Data.ByteString.Lazy as B import qualified Data.ByteString.Lazy.Char8 as B8 -import Text.Pandoc.UTF8 ( fromStringLazy, toString ) +import qualified Text.Pandoc.UTF8 as UTF8 import Text.Pandoc.SelfContained ( makeSelfContained ) import Codec.Archive.Zip +import Control.Applicative ((<$>)) import Data.Time.Clock.POSIX import Data.Time import System.Locale @@ -54,7 +56,7 @@ import Text.XML.Light hiding (ppTopElement) import Text.Pandoc.UUID import Text.Pandoc.Writers.HTML import Text.Pandoc.Writers.Markdown ( writePlain ) -import Data.Char ( toLower ) +import Data.Char ( toLower, isDigit ) import Network.URI ( unEscapeString ) import Text.Pandoc.MIME (getMimeType) #if MIN_VERSION_base(4,6,0) @@ -70,12 +72,237 @@ import Text.Blaze.Html.Renderer.Utf8 (renderHtml) -- in filenames, chapter0003.xhtml. data Chapter = Chapter (Maybe [Int]) [Block] +data EPUBMetadata = EPUBMetadata{ + epubIdentifier :: [Identifier] + , epubTitle :: [Title] + , epubDate :: String + , epubLanguage :: String + , epubCreator :: [Creator] + , epubContributor :: [Creator] + , epubSubject :: [String] + , epubDescription :: Maybe String + , epubType :: Maybe String + , epubFormat :: Maybe String + , epubPublisher :: Maybe String + , epubSource :: Maybe String + , epubRelation :: Maybe String + , epubCoverage :: Maybe String + , epubRights :: Maybe String + , epubCoverImage :: Maybe String + , epubStylesheet :: Maybe Stylesheet + } deriving Show + +data Stylesheet = StylesheetPath FilePath + | StylesheetContents String + deriving Show + +data Creator = Creator{ + creatorText :: String + , creatorRole :: Maybe String + , creatorFileAs :: Maybe String + } deriving Show + +data Identifier = Identifier{ + identifierText :: String + , identifierScheme :: Maybe String + } deriving Show + +data Title = Title{ + titleText :: String + , titleFileAs :: Maybe String + , titleType :: Maybe String + } deriving Show + +dcName :: String -> QName +dcName n = QName n Nothing (Just "dc") + +dcNode :: Node t => String -> t -> Element +dcNode = node . dcName + +opfName :: String -> QName +opfName n = QName n Nothing (Just "opf") + +plainify :: [Inline] -> String +plainify t = + trimr $ writePlain def{ writerStandalone = False } + $ Pandoc nullMeta [Plain $ walk removeNote t] + +removeNote :: Inline -> Inline +removeNote (Note _) = Str "" +removeNote x = x + +getEPUBMetadata :: WriterOptions -> Meta -> IO EPUBMetadata +getEPUBMetadata opts meta = do + let md = metadataFromMeta opts meta + let elts = onlyElems $ parseXML $ writerEpubMetadata opts + let md' = foldr addMetadataFromXML md elts + let addIdentifier m = + if null (epubIdentifier m) + then do + randomId <- fmap show getRandomUUID + return $ m{ epubIdentifier = [Identifier randomId Nothing] } + else return m + let addLanguage m = + if null (epubLanguage m) + then case lookup "lang" (writerVariables opts) of + Just x -> return m{ epubLanguage = x } + Nothing -> do + localeLang <- catch (liftM + (map (\c -> if c == '_' then '-' else c) . + takeWhile (/='.')) $ getEnv "LANG") + (\e -> let _ = (e :: SomeException) in return "en-US") + return m{ epubLanguage = localeLang } + else return m + let fixDate m = + if null (epubDate m) + then do + currentTime <- getCurrentTime + return $ m{ epubDate = showDateTimeISO8601 currentTime } + else return m + let addAuthor m = + if any (\c -> creatorRole c == Just "aut") $ epubCreator m + then return m + else do + let authors' = map plainify $ docAuthors meta + let toAuthor name = Creator{ creatorText = name + , creatorRole = Just "aut" + , creatorFileAs = Nothing } + return $ m{ epubCreator = map toAuthor authors' ++ epubCreator m } + addIdentifier md' >>= fixDate >>= addAuthor >>= addLanguage + +addMetadataFromXML :: Element -> EPUBMetadata -> EPUBMetadata +addMetadataFromXML e@(Element (QName name _ (Just "dc")) attrs _ _) md + | name == "identifier" = md{ epubIdentifier = + Identifier{ identifierText = strContent e + , identifierScheme = lookupAttr (opfName "scheme") attrs + } : epubIdentifier md } + | name == "title" = md{ epubTitle = + Title{ titleText = strContent e + , titleFileAs = getAttr "file-as" + , titleType = getAttr "type" + } : epubTitle md } + | name == "date" = md{ epubDate = fromMaybe "" $ normalizeDate' + $ strContent e } + | name == "language" = md{ epubLanguage = strContent e } + | name == "creator" = md{ epubCreator = + Creator{ creatorText = strContent e + , creatorRole = getAttr "role" + , creatorFileAs = getAttr "file-as" + } : epubCreator md } + | name == "contributor" = md{ epubContributor = + Creator { creatorText = strContent e + , creatorRole = getAttr "role" + , creatorFileAs = getAttr "file-as" + } : epubContributor md } + | name == "subject" = md{ epubSubject = strContent e : epubSubject md } + | name == "description" = md { epubDescription = Just $ strContent e } + | name == "type" = md { epubType = Just $ strContent e } + | name == "format" = md { epubFormat = Just $ strContent e } + | name == "type" = md { epubType = Just $ strContent e } + | name == "publisher" = md { epubPublisher = Just $ strContent e } + | name == "source" = md { epubSource = Just $ strContent e } + | name == "relation" = md { epubRelation = Just $ strContent e } + | name == "coverage" = md { epubCoverage = Just $ strContent e } + | name == "rights" = md { epubRights = Just $ strContent e } + | otherwise = md + where getAttr n = lookupAttr (opfName n) attrs +addMetadataFromXML _ md = md + +metaValueToString :: MetaValue -> String +metaValueToString (MetaString s) = s +metaValueToString (MetaInlines ils) = plainify ils +metaValueToString (MetaBlocks bs) = plainify $ query (:[]) bs +metaValueToString (MetaBool b) = show b +metaValueToString _ = "" + +getList :: String -> Meta -> (MetaValue -> a) -> [a] +getList s meta handleMetaValue = + case lookupMeta s meta of + Just (MetaList xs) -> map handleMetaValue xs + Just mv -> [handleMetaValue mv] + Nothing -> [] + +getIdentifier :: Meta -> [Identifier] +getIdentifier meta = getList "identifier" meta handleMetaValue + where handleMetaValue (MetaMap m) = + Identifier{ identifierText = maybe "" metaValueToString + $ M.lookup "text" m + , identifierScheme = metaValueToString <$> + M.lookup "scheme" m } + handleMetaValue mv = Identifier (metaValueToString mv) Nothing + +getTitle :: Meta -> [Title] +getTitle meta = getList "title" meta handleMetaValue + where handleMetaValue (MetaMap m) = + Title{ titleText = maybe "" metaValueToString $ M.lookup "text" m + , titleFileAs = metaValueToString <$> M.lookup "file-as" m + , titleType = metaValueToString <$> M.lookup "type" m } + handleMetaValue mv = Title (metaValueToString mv) Nothing Nothing + +getCreator :: String -> Meta -> [Creator] +getCreator s meta = getList s meta handleMetaValue + where handleMetaValue (MetaMap m) = + Creator{ creatorText = maybe "" metaValueToString $ M.lookup "text" m + , creatorFileAs = metaValueToString <$> M.lookup "file-as" m + , creatorRole = metaValueToString <$> M.lookup "role" m } + handleMetaValue mv = Creator (metaValueToString mv) Nothing Nothing + +simpleList :: String -> Meta -> [String] +simpleList s meta = + case lookupMeta s meta of + Just (MetaList xs) -> map metaValueToString xs + Just x -> [metaValueToString x] + Nothing -> [] + +metadataFromMeta :: WriterOptions -> Meta -> EPUBMetadata +metadataFromMeta opts meta = EPUBMetadata{ + epubIdentifier = identifiers + , epubTitle = titles + , epubDate = date + , epubLanguage = language + , epubCreator = creators + , epubContributor = contributors + , epubSubject = subjects + , epubDescription = description + , epubType = epubtype + , epubFormat = format + , epubPublisher = publisher + , epubSource = source + , epubRelation = relation + , epubCoverage = coverage + , epubRights = rights + , epubCoverImage = coverImage + , epubStylesheet = stylesheet + } + where identifiers = getIdentifier meta + titles = getTitle meta + date = fromMaybe "" $ + (metaValueToString <$> lookupMeta "date" meta) >>= normalizeDate' + language = maybe "" metaValueToString $ + lookupMeta "language" meta `mplus` lookupMeta "lang" meta + creators = getCreator "creator" meta + contributors = getCreator "contributor" meta + subjects = simpleList "subject" meta + description = metaValueToString <$> lookupMeta "description" meta + epubtype = metaValueToString <$> lookupMeta "type" meta + format = metaValueToString <$> lookupMeta "format" meta + publisher = metaValueToString <$> lookupMeta "publisher" meta + source = metaValueToString <$> lookupMeta "source" meta + relation = metaValueToString <$> lookupMeta "relation" meta + coverage = metaValueToString <$> lookupMeta "coverage" meta + rights = metaValueToString <$> lookupMeta "rights" meta + coverImage = lookup "epub-cover-image" (writerVariables opts) `mplus` + (metaValueToString <$> lookupMeta "cover-image" meta) + stylesheet = (StylesheetContents <$> writerEpubStylesheet opts) `mplus` + ((StylesheetPath . metaValueToString) <$> + lookupMeta "stylesheet" meta) + -- | Produce an EPUB file from a Pandoc document. writeEPUB :: WriterOptions -- ^ Writer options -> Pandoc -- ^ Document to convert -> IO B.ByteString writeEPUB opts doc@(Pandoc meta _) = do - let version = maybe EPUB2 id (writerEpubVersion opts) + let version = fromMaybe EPUB2 (writerEpubVersion opts) let epub3 = version == EPUB3 epochtime <- floor `fmap` getPOSIXTime let mkEntry path content = toEntry path epochtime content @@ -93,11 +320,11 @@ writeEPUB opts doc@(Pandoc meta _) = do then MathML Nothing else writerHTMLMathMethod opts , writerWrapText = False } - let mbCoverImage = lookup "epub-cover-image" vars + metadata <- getEPUBMetadata opts' meta -- cover page (cpgEntry, cpicEntry) <- - case mbCoverImage of + case epubCoverImage metadata of Nothing -> return ([],[]) Just img -> do let coverImage = "cover-image" ++ takeExtension img @@ -179,32 +406,33 @@ writeEPUB opts doc@(Pandoc meta _) = do chapToEntry num (Chapter mbnum bs) = mkEntry (showChapter num) $ renderHtml $ writeHtml opts'{ writerNumberOffset = - maybe [] id mbnum } + fromMaybe [] mbnum } $ case bs of (Header _ _ xs : _) -> - Pandoc (setMeta "title" (fromList xs) nullMeta) bs + -- remove notes or we get doubled footnotes + Pandoc (setMeta "title" (walk removeNote $ fromList xs) + nullMeta) bs _ -> Pandoc nullMeta bs let chapterEntries = zipWith chapToEntry [1..] chapters -- incredibly inefficient (TODO): - let containsMathML ent = "<math" `isInfixOf` (B8.unpack $ fromEntry ent) + let containsMathML ent = epub3 && + "<math" `isInfixOf` (B8.unpack $ fromEntry ent) + let containsSVG ent = epub3 && + "<svg" `isInfixOf` (B8.unpack $ fromEntry ent) + let props ent = ["mathml" | containsMathML ent] ++ ["svg" | containsSVG ent] -- contents.opf - localeLang <- catch (liftM (map (\c -> if c == '_' then '-' else c) . - takeWhile (/='.')) $ getEnv "LANG") - (\e -> let _ = (e :: SomeException) in return "en-US") - let lang = case lookup "lang" (writerVariables opts') of - Just x -> x - Nothing -> localeLang - uuid <- getRandomUUID let chapterNode ent = unode "item" ! ([("id", takeBaseName $ eRelativePath ent), ("href", eRelativePath ent), ("media-type", "application/xhtml+xml")] - ++ [("properties","mathml") | epub3 && - containsMathML ent]) $ () + ++ case props ent of + [] -> [] + xs -> [("properties", unwords xs)]) + $ () let chapterRefNode ent = unode "itemref" ! [("idref", takeBaseName $ eRelativePath ent)] $ () let pictureNode ent = unode "item" ! @@ -215,23 +443,23 @@ writeEPUB opts doc@(Pandoc meta _) = do let fontNode ent = unode "item" ! [("id", takeBaseName $ eRelativePath ent), ("href", eRelativePath ent), - ("media-type", maybe "" id $ getMimeType $ eRelativePath ent)] $ () - let plainify t = trimr $ - writePlain opts'{ writerStandalone = False } $ - Pandoc meta [Plain t] - let plainTitle = plainify $ docTitle meta - let plainAuthors = map plainify $ docAuthors meta + ("media-type", fromMaybe "" $ getMimeType $ eRelativePath ent)] $ () + let plainTitle = case docTitle meta of + [] -> case epubTitle metadata of + [] -> "UNTITLED" + (x:_) -> titleText x + x -> plainify x + let uuid = case epubIdentifier metadata of + (x:_) -> identifierText x -- use first identifier as UUID + [] -> error "epubIdentifier is null" -- shouldn't happen currentTime <- getCurrentTime - let plainDate = maybe (showDateTimeISO8601 currentTime) id - $ normalizeDate $ stringify $ docDate meta - let contentsData = fromStringLazy $ ppTopElement $ + let contentsData = UTF8.fromStringLazy $ ppTopElement $ unode "package" ! [("version", case version of EPUB2 -> "2.0" EPUB3 -> "3.0") ,("xmlns","http://www.idpf.org/2007/opf") - ,("unique-identifier","BookId")] $ - [ metadataElement version (writerEpubMetadata opts') - uuid lang plainTitle plainAuthors plainDate currentTime mbCoverImage + ,("unique-identifier","epub-id-1")] $ + [ metadataElement version metadata currentTime , unode "manifest" $ [ unode "item" ! [("id","ncx"), ("href","toc.ncx") ,("media-type","application/x-dtbncx+xml")] $ () @@ -243,10 +471,15 @@ writeEPUB opts doc@(Pandoc meta _) = do [("properties","nav") | epub3 ]) $ () ] ++ map chapterNode (cpgEntry ++ (tpEntry : chapterEntries)) ++ - map pictureNode (cpicEntry ++ picEntries) ++ + (case cpicEntry of + [] -> [] + (x:_) -> [add_attrs + [Attr (unqual "properties") "cover-image" | epub3] + (pictureNode x)]) ++ + map pictureNode picEntries ++ map fontNode fontEntries , unode "spine" ! [("toc","ncx")] $ - case mbCoverImage of + case epubCoverImage metadata of Nothing -> [] Just _ -> [ unode "itemref" ! [("idref", "cover"),("linear","no")] $ () ] @@ -260,8 +493,13 @@ writeEPUB opts doc@(Pandoc meta _) = do else "no")] $ ()) : map chapterRefNode chapterEntries) , unode "guide" $ - unode "reference" ! - [("type","toc"),("title",plainTitle),("href","nav.xhtml")] $ () + [ unode "reference" ! + [("type","toc"),("title",plainTitle), + ("href","nav.xhtml")] $ () + ] ++ + [ unode "reference" ! + [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () | epubCoverImage metadata /= Nothing + ] ] let contentsEntry = mkEntry "content.opf" contentsData @@ -303,19 +541,19 @@ writeEPUB opts doc@(Pandoc meta _) = do [ unode "navLabel" $ unode "text" (plainify $ docTitle meta) , unode "content" ! [("src","title_page.xhtml")] $ () ] - let tocData = fromStringLazy $ ppTopElement $ + let tocData = UTF8.fromStringLazy $ ppTopElement $ unode "ncx" ! [("version","2005-1") ,("xmlns","http://www.daisy.org/z3986/2005/ncx/")] $ [ unode "head" $ [ unode "meta" ! [("name","dtb:uid") - ,("content", show uuid)] $ () + ,("content", uuid)] $ () , unode "meta" ! [("name","dtb:depth") ,("content", "1")] $ () , unode "meta" ! [("name","dtb:totalPageCount") ,("content", "0")] $ () , unode "meta" ! [("name","dtb:maxPageNumber") ,("content", "0")] $ () - ] ++ case mbCoverImage of + ] ++ case epubCoverImage metadata of Nothing -> [] Just _ -> [unode "meta" ! [("name","cover"), ("content","cover-image")] $ ()] @@ -335,7 +573,7 @@ writeEPUB opts doc@(Pandoc meta _) = do (_:_) -> [unode "ol" ! [("class","toc")] $ subs] let navtag = if epub3 then "nav" else "div" - let navData = fromStringLazy $ ppTopElement $ + let navData = UTF8.fromStringLazy $ ppTopElement $ unode "html" ! [("xmlns","http://www.w3.org/1999/xhtml") ,("xmlns:epub","http://www.idpf.org/2007/ops")] $ [ unode "head" $ @@ -349,10 +587,10 @@ writeEPUB opts doc@(Pandoc meta _) = do let navEntry = mkEntry "nav.xhtml" navData -- mimetype - let mimetypeEntry = mkEntry "mimetype" $ fromStringLazy "application/epub+zip" + let mimetypeEntry = mkEntry "mimetype" $ UTF8.fromStringLazy "application/epub+zip" -- container.xml - let containerData = fromStringLazy $ ppTopElement $ + let containerData = UTF8.fromStringLazy $ ppTopElement $ unode "container" ! [("version","1.0") ,("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")] $ unode "rootfiles" $ @@ -361,18 +599,19 @@ writeEPUB opts doc@(Pandoc meta _) = do let containerEntry = mkEntry "META-INF/container.xml" containerData -- com.apple.ibooks.display-options.xml - let apple = fromStringLazy $ ppTopElement $ + let apple = UTF8.fromStringLazy $ ppTopElement $ unode "display_options" $ unode "platform" ! [("name","*")] $ unode "option" ! [("name","specified-fonts")] $ "true" let appleEntry = mkEntry "META-INF/com.apple.ibooks.display-options.xml" apple -- stylesheet - stylesheet <- case writerEpubStylesheet opts of - Just s -> return s - Nothing -> toString `fmap` + stylesheet <- case epubStylesheet metadata of + Just (StylesheetPath fp) -> UTF8.readFile fp + Just (StylesheetContents s) -> return s + Nothing -> UTF8.toString `fmap` readDataFile (writerUserDataDir opts) "epub.css" - let stylesheetEntry = mkEntry "stylesheet.css" $ fromStringLazy stylesheet + let stylesheetEntry = mkEntry "stylesheet.css" $ UTF8.fromStringLazy stylesheet -- construct archive let archive = foldr addEntryToArchive emptyArchive @@ -381,33 +620,97 @@ writeEPUB opts doc@(Pandoc meta _) = do (picEntries ++ cpicEntry ++ cpgEntry ++ chapterEntries ++ fontEntries)) return $ fromArchive archive -metadataElement :: EPUBVersion -> String -> UUID -> String -> String -> [String] - -> String -> UTCTime -> Maybe a -> Element -metadataElement version metadataXML uuid lang title authors date currentTime mbCoverImage = - let userNodes = parseXML metadataXML - elt = unode "metadata" ! [("xmlns:dc","http://purl.org/dc/elements/1.1/") - ,("xmlns:opf","http://www.idpf.org/2007/opf")] $ - filter isMetadataElement $ onlyElems userNodes - dublinElements = ["contributor","coverage","creator","date", - "description","format","identifier","language","publisher", - "relation","rights","source","subject","title","type"] - isMetadataElement e = (qPrefix (elName e) == Just "dc" && - qName (elName e) `elem` dublinElements) || - (qPrefix (elName e) == Nothing && - qName (elName e) `elem` ["link","meta"]) - contains e n = not (null (findElements (QName n Nothing (Just "dc")) e)) - newNodes = [ unode "dc:title" title | not (elt `contains` "title") ] ++ - [ unode "dc:language" lang | not (elt `contains` "language") ] ++ - [ unode "dc:identifier" ! [("id","BookId")] $ show uuid | - not (elt `contains` "identifier") ] ++ - [ unode "dc:creator" ! [("opf:role","aut") | version == EPUB2] - $ a | a <- authors, not (elt `contains` "creator") ] ++ - [ unode "dc:date" date | not (elt `contains` "date") ] ++ - [ unode "meta" ! [("property", "dcterms:modified")] $ - (showDateTimeISO8601 currentTime) | version == EPUB3] ++ - [ unode "meta" ! [("name","cover"), ("content","cover-image")] $ () | - not (isNothing mbCoverImage) ] - in elt{ elContent = elContent elt ++ map Elem newNodes } +metadataElement :: EPUBVersion -> EPUBMetadata -> UTCTime -> Element +metadataElement version md currentTime = + unode "metadata" ! [("xmlns:dc","http://purl.org/dc/elements/1.1/") + ,("xmlns:opf","http://www.idpf.org/2007/opf")] $ mdNodes + where mdNodes = identifierNodes ++ titleNodes ++ dateNodes ++ languageNodes + ++ creatorNodes ++ contributorNodes ++ subjectNodes + ++ descriptionNodes ++ typeNodes ++ formatNodes + ++ publisherNodes ++ sourceNodes ++ relationNodes + ++ coverageNodes ++ rightsNodes ++ coverImageNodes + ++ modifiedNodes + withIds base f = concat . zipWith f (map (\x -> base ++ ('-' : show x)) + ([1..] :: [Int])) + identifierNodes = withIds "epub-id" toIdentifierNode $ + epubIdentifier md + titleNodes = withIds "epub-title" toTitleNode $ epubTitle md + dateNodes = dcTag' "date" $ epubDate md + languageNodes = [dcTag "language" $ epubLanguage md] + creatorNodes = withIds "epub-creator" (toCreatorNode "creator") $ + epubCreator md + contributorNodes = withIds "epub-contributor" + (toCreatorNode "contributor") $ epubContributor md + subjectNodes = map (dcTag "subject") $ epubSubject md + descriptionNodes = maybe [] (dcTag' "description") $ epubDescription md + typeNodes = maybe [] (dcTag' "type") $ epubType md + formatNodes = maybe [] (dcTag' "format") $ epubFormat md + publisherNodes = maybe [] (dcTag' "publisher") $ epubPublisher md + sourceNodes = maybe [] (dcTag' "source") $ epubSource md + relationNodes = maybe [] (dcTag' "relation") $ epubRelation md + coverageNodes = maybe [] (dcTag' "coverage") $ epubCoverage md + rightsNodes = maybe [] (dcTag' "rights") $ epubRights md + coverImageNodes = maybe [] + (const $ [unode "meta" ! [("name","cover"), + ("content","cover-image")] $ ()]) + $ epubCoverImage md + modifiedNodes = [ unode "meta" ! [("property", "dcterms:modified")] $ + (showDateTimeISO8601 currentTime) | version == EPUB3 ] + dcTag n s = unode ("dc:" ++ n) s + dcTag' n s = [dcTag n s] + toIdentifierNode id' (Identifier txt scheme) + | version == EPUB2 = [dcNode "identifier" ! + ([("id",id')] ++ maybe [] (\x -> [("opf:scheme", x)]) scheme) $ + txt] + | otherwise = [dcNode "identifier" ! [("id",id')] $ txt] ++ + maybe [] (\x -> [unode "meta" ! + [("refines",'#':id'),("property","identifier-type"), + ("scheme","onix:codelist5")] $ x]) + (schemeToOnix `fmap` scheme) + toCreatorNode s id' creator + | version == EPUB2 = [dcNode s ! + ([("id",id')] ++ + maybe [] (\x -> [("opf:file-as",x)]) (creatorFileAs creator) ++ + maybe [] (\x -> [("opf:role",x)]) + (creatorRole creator >>= toRelator)) $ creatorText creator] + | otherwise = [dcNode s ! [("id",id')] $ creatorText creator] ++ + maybe [] (\x -> [unode "meta" ! + [("refines",'#':id'),("property","file-as")] $ x]) + (creatorFileAs creator) ++ + maybe [] (\x -> [unode "meta" ! + [("refines",'#':id'),("property","role"), + ("scheme","marc:relators")] $ x]) + (creatorRole creator >>= toRelator) + toTitleNode id' title + | version == EPUB2 = [dcNode "title" ! + ([("id",id')] ++ + maybe [] (\x -> [("opf:file-as",x)]) (titleFileAs title) ++ + maybe [] (\x -> [("opf:title-type",x)]) (titleType title)) $ + titleText title] + | otherwise = [dcNode "title" ! [("id",id')] $ titleText title] + ++ + maybe [] (\x -> [unode "meta" ! + [("refines",'#':id'),("property","file-as")] $ x]) + (titleFileAs title) ++ + maybe [] (\x -> [unode "meta" ! + [("refines",'#':id'),("property","title-type")] $ x]) + (titleType title) + schemeToOnix "ISBN-10" = "02" + schemeToOnix "GTIN-13" = "03" + schemeToOnix "UPC" = "04" + schemeToOnix "ISMN-10" = "05" + schemeToOnix "DOI" = "06" + schemeToOnix "LCCN" = "13" + schemeToOnix "GTIN-14" = "14" + schemeToOnix "ISBN-13" = "15" + schemeToOnix "Legal deposit number" = "17" + schemeToOnix "URN" = "22" + schemeToOnix "OCLC" = "23" + schemeToOnix "ISMN-13" = "25" + schemeToOnix "ISBN-A" = "26" + schemeToOnix "JP" = "27" + schemeToOnix "OLCC" = "28" + schemeToOnix _ = "01" showDateTimeISO8601 :: UTCTime -> String showDateTimeISO8601 = formatTime defaultTimeLocale "%FT%TZ" @@ -457,15 +760,9 @@ ppTopElement = ("<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n" ++) . unEntity . unEntity (x:xs) = x : unEntity xs imageTypeOf :: FilePath -> Maybe String -imageTypeOf x = case drop 1 (map toLower (takeExtension x)) of - "jpg" -> Just "image/jpeg" - "jpeg" -> Just "image/jpeg" - "jfif" -> Just "image/jpeg" - "png" -> Just "image/png" - "gif" -> Just "image/gif" - "svg" -> Just "image/svg+xml" - _ -> Nothing - +imageTypeOf x = case getMimeType x of + Just y@('i':'m':'a':'g':'e':_) -> Just y + _ -> Nothing data IdentState = IdentState{ chapterNumber :: Int, @@ -519,3 +816,287 @@ replaceRefs refTable = walk replaceOneRef Just url -> Link lab (url,tit) Nothing -> x replaceOneRef x = x + +-- Variant of normalizeDate that allows partial dates: YYYY, YYYY-MM +normalizeDate' :: String -> Maybe String +normalizeDate' xs = + let xs' = trim xs in + case xs' of + [y1,y2,y3,y4] | all isDigit [y1,y2,y3,y4] -> Just xs' -- YYYY + [y1,y2,y3,y4,'-',m1,m2] | all isDigit [y1,y2,y3,y4,m1,m2] -- YYYY-MM + -> Just xs' + _ -> normalizeDate xs' + +toRelator :: String -> Maybe String +toRelator x + | x `elem` relators = Just x + | otherwise = lookup (map toLower x) relatorMap + +relators :: [String] +relators = map snd relatorMap + +relatorMap :: [(String, String)] +relatorMap = + [("abridger", "abr") + ,("actor", "act") + ,("adapter", "adp") + ,("addressee", "rcp") + ,("analyst", "anl") + ,("animator", "anm") + ,("annotator", "ann") + ,("appellant", "apl") + ,("appellee", "ape") + ,("applicant", "app") + ,("architect", "arc") + ,("arranger", "arr") + ,("art copyist", "acp") + ,("art director", "adi") + ,("artist", "art") + ,("artistic director", "ard") + ,("assignee", "asg") + ,("associated name", "asn") + ,("attributed name", "att") + ,("auctioneer", "auc") + ,("author", "aut") + ,("author in quotations or text abstracts", "aqt") + ,("author of afterword, colophon, etc.", "aft") + ,("author of dialog", "aud") + ,("author of introduction, etc.", "aui") + ,("autographer", "ato") + ,("bibliographic antecedent", "ant") + ,("binder", "bnd") + ,("binding designer", "bdd") + ,("blurb writer", "blw") + ,("book designer", "bkd") + ,("book producer", "bkp") + ,("bookjacket designer", "bjd") + ,("bookplate designer", "bpd") + ,("bookseller", "bsl") + ,("braille embosser", "brl") + ,("broadcaster", "brd") + ,("calligrapher", "cll") + ,("cartographer", "ctg") + ,("caster", "cas") + ,("censor", "cns") + ,("choreographer", "chr") + ,("cinematographer", "cng") + ,("client", "cli") + ,("collection registrar", "cor") + ,("collector", "col") + ,("collotyper", "clt") + ,("colorist", "clr") + ,("commentator", "cmm") + ,("commentator for written text", "cwt") + ,("compiler", "com") + ,("complainant", "cpl") + ,("complainant-appellant", "cpt") + ,("complainant-appellee", "cpe") + ,("composer", "cmp") + ,("compositor", "cmt") + ,("conceptor", "ccp") + ,("conductor", "cnd") + ,("conservator", "con") + ,("consultant", "csl") + ,("consultant to a project", "csp") + ,("contestant", "cos") + ,("contestant-appellant", "cot") + ,("contestant-appellee", "coe") + ,("contestee", "cts") + ,("contestee-appellant", "ctt") + ,("contestee-appellee", "cte") + ,("contractor", "ctr") + ,("contributor", "ctb") + ,("copyright claimant", "cpc") + ,("copyright holder", "cph") + ,("corrector", "crr") + ,("correspondent", "crp") + ,("costume designer", "cst") + ,("court governed", "cou") + ,("court reporter", "crt") + ,("cover designer", "cov") + ,("creator", "cre") + ,("curator", "cur") + ,("dancer", "dnc") + ,("data contributor", "dtc") + ,("data manager", "dtm") + ,("dedicatee", "dte") + ,("dedicator", "dto") + ,("defendant", "dfd") + ,("defendant-appellant", "dft") + ,("defendant-appellee", "dfe") + ,("degree granting institution", "dgg") + ,("delineator", "dln") + ,("depicted", "dpc") + ,("depositor", "dpt") + ,("designer", "dsr") + ,("director", "drt") + ,("dissertant", "dis") + ,("distribution place", "dbp") + ,("distributor", "dst") + ,("donor", "dnr") + ,("draftsman", "drm") + ,("dubious author", "dub") + ,("editor", "edt") + ,("editor of compilation", "edc") + ,("editor of moving image work", "edm") + ,("electrician", "elg") + ,("electrotyper", "elt") + ,("enacting jurisdiction", "enj") + ,("engineer", "eng") + ,("engraver", "egr") + ,("etcher", "etr") + ,("event place", "evp") + ,("expert", "exp") + ,("facsimilist", "fac") + ,("field director", "fld") + ,("film director", "fmd") + ,("film distributor", "fds") + ,("film editor", "flm") + ,("film producer", "fmp") + ,("filmmaker", "fmk") + ,("first party", "fpy") + ,("forger", "frg") + ,("former owner", "fmo") + ,("funder", "fnd") + ,("geographic information specialist", "gis") + ,("honoree", "hnr") + ,("host", "hst") + ,("host institution", "his") + ,("illuminator", "ilu") + ,("illustrator", "ill") + ,("inscriber", "ins") + ,("instrumentalist", "itr") + ,("interviewee", "ive") + ,("interviewer", "ivr") + ,("inventor", "inv") + ,("issuing body", "isb") + ,("judge", "jud") + ,("jurisdiction governed", "jug") + ,("laboratory", "lbr") + ,("laboratory director", "ldr") + ,("landscape architect", "lsa") + ,("lead", "led") + ,("lender", "len") + ,("libelant", "lil") + ,("libelant-appellant", "lit") + ,("libelant-appellee", "lie") + ,("libelee", "lel") + ,("libelee-appellant", "let") + ,("libelee-appellee", "lee") + ,("librettist", "lbt") + ,("licensee", "lse") + ,("licensor", "lso") + ,("lighting designer", "lgd") + ,("lithographer", "ltg") + ,("lyricist", "lyr") + ,("manufacture place", "mfp") + ,("manufacturer", "mfr") + ,("marbler", "mrb") + ,("markup editor", "mrk") + ,("metadata contact", "mdc") + ,("metal-engraver", "mte") + ,("moderator", "mod") + ,("monitor", "mon") + ,("music copyist", "mcp") + ,("musical director", "msd") + ,("musician", "mus") + ,("narrator", "nrt") + ,("onscreen presenter", "osp") + ,("opponent", "opn") + ,("organizer of meeting", "orm") + ,("originator", "org") + ,("other", "oth") + ,("owner", "own") + ,("panelist", "pan") + ,("papermaker", "ppm") + ,("patent applicant", "pta") + ,("patent holder", "pth") + ,("patron", "pat") + ,("performer", "prf") + ,("permitting agency", "pma") + ,("photographer", "pht") + ,("plaintiff", "ptf") + ,("plaintiff-appellant", "ptt") + ,("plaintiff-appellee", "pte") + ,("platemaker", "plt") + ,("praeses", "pra") + ,("presenter", "pre") + ,("printer", "prt") + ,("printer of plates", "pop") + ,("printmaker", "prm") + ,("process contact", "prc") + ,("producer", "pro") + ,("production company", "prn") + ,("production designer", "prs") + ,("production manager", "pmn") + ,("production personnel", "prd") + ,("production place", "prp") + ,("programmer", "prg") + ,("project director", "pdr") + ,("proofreader", "pfr") + ,("provider", "prv") + ,("publication place", "pup") + ,("publisher", "pbl") + ,("publishing director", "pbd") + ,("puppeteer", "ppt") + ,("radio director", "rdd") + ,("radio producer", "rpc") + ,("recording engineer", "rce") + ,("recordist", "rcd") + ,("redaktor", "red") + ,("renderer", "ren") + ,("reporter", "rpt") + ,("repository", "rps") + ,("research team head", "rth") + ,("research team member", "rtm") + ,("researcher", "res") + ,("respondent", "rsp") + ,("respondent-appellant", "rst") + ,("respondent-appellee", "rse") + ,("responsible party", "rpy") + ,("restager", "rsg") + ,("restorationist", "rsr") + ,("reviewer", "rev") + ,("rubricator", "rbr") + ,("scenarist", "sce") + ,("scientific advisor", "sad") + ,("screenwriter", "aus") + ,("scribe", "scr") + ,("sculptor", "scl") + ,("second party", "spy") + ,("secretary", "sec") + ,("seller", "sll") + ,("set designer", "std") + ,("setting", "stg") + ,("signer", "sgn") + ,("singer", "sng") + ,("sound designer", "sds") + ,("speaker", "spk") + ,("sponsor", "spn") + ,("stage director", "sgd") + ,("stage manager", "stm") + ,("standards body", "stn") + ,("stereotyper", "str") + ,("storyteller", "stl") + ,("supporting host", "sht") + ,("surveyor", "srv") + ,("teacher", "tch") + ,("technical director", "tcd") + ,("television director", "tld") + ,("television producer", "tlp") + ,("thesis advisor", "ths") + ,("transcriber", "trc") + ,("translator", "trl") + ,("type designer", "tyd") + ,("typographer", "tyg") + ,("university place", "uvp") + ,("videographer", "vdg") + ,("witness", "wit") + ,("wood engraver", "wde") + ,("woodcutter", "wdc") + ,("writer of accompanying material", "wam") + ,("writer of added commentary", "wac") + ,("writer of added lyrics", "wal") + ,("writer of added text", "wat") + ] + diff --git a/src/Text/Pandoc/Writers/FB2.hs b/src/Text/Pandoc/Writers/FB2.hs index adbe948be..803617f95 100644 --- a/src/Text/Pandoc/Writers/FB2.hs +++ b/src/Text/Pandoc/Writers/FB2.hs @@ -44,7 +44,7 @@ import qualified Text.XML.Light.Cursor as XC import Text.Pandoc.Definition import Text.Pandoc.Options (WriterOptions(..), HTMLMathMethod(..), def) -import Text.Pandoc.Shared (orderedListMarkers) +import Text.Pandoc.Shared (orderedListMarkers, isHeaderBlock) import Text.Pandoc.Walk -- | Data to be written at the end of the document: @@ -157,9 +157,7 @@ renderSection level (ttl, body) = do else cMapM blockToXml body return $ el "section" (title ++ content) where - hasSubsections = any isHeader - isHeader (Header _ _ _) = True - isHeader _ = False + hasSubsections = any isHeaderBlock -- | Only <p> and <empty-line> are allowed within <title> in FB2. formatTitle :: [Inline] -> [Content] diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index c1cca291b..805bb57f1 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -45,7 +45,7 @@ import Numeric ( showHex ) import Data.Char ( ord, toLower ) import Data.List ( isPrefixOf, intersperse ) import Data.String ( fromString ) -import Data.Maybe ( catMaybes ) +import Data.Maybe ( catMaybes, fromMaybe ) import Control.Monad.State import Text.Blaze.Html hiding(contents) import Text.Blaze.Internal(preEscapedString) @@ -118,7 +118,7 @@ pandocToHtml opts (Pandoc meta blocks) = do let stringifyHTML = escapeStringForXML . stringify let authsMeta = map stringifyHTML $ docAuthors meta let dateMeta = stringifyHTML $ docDate meta - let slideLevel = maybe (getSlideLevel blocks) id $ writerSlideLevel opts + let slideLevel = fromMaybe (getSlideLevel blocks) $ writerSlideLevel opts let sects = hierarchicalize $ if writerSlideVariant opts == NoSlides then blocks @@ -475,28 +475,21 @@ blockToHtml opts (BlockQuote blocks) = else do contents <- blockListToHtml opts blocks return $ H.blockquote $ nl opts >> contents >> nl opts -blockToHtml opts (Header level (ident,_,_) lst) = do +blockToHtml opts (Header level (_,_,_) lst) = do contents <- inlineListToHtml opts lst secnum <- liftM stSecNum get let contents' = if writerNumberSections opts && not (null secnum) then (H.span ! A.class_ "header-section-number" $ toHtml $ showSecNum secnum) >> strToHtml " " >> contents else contents - let revealSlash = ['/' | writerSlideVariant opts == RevealJsSlides] - let contents'' = if writerTableOfContents opts && not (null ident) - then H.a ! A.href (toValue $ - '#' : revealSlash ++ - writerIdentifierPrefix opts ++ - ident) $ contents' - else contents' return $ case level of - 1 -> H.h1 contents'' - 2 -> H.h2 contents'' - 3 -> H.h3 contents'' - 4 -> H.h4 contents'' - 5 -> H.h5 contents'' - 6 -> H.h6 contents'' - _ -> H.p contents'' + 1 -> H.h1 contents' + 2 -> H.h2 contents' + 3 -> H.h3 contents' + 4 -> H.h4 contents' + 5 -> H.h5 contents' + 6 -> H.h6 contents' + _ -> H.p contents' blockToHtml opts (BulletList lst) = do contents <- mapM (blockListToHtml opts) lst return $ unordList opts contents @@ -524,7 +517,7 @@ blockToHtml opts (DefinitionList lst) = do contents <- mapM (\(term, defs) -> do term' <- if null term then return mempty - else liftM (H.dt) $ inlineListToHtml opts term + else liftM H.dt $ inlineListToHtml opts term defs' <- mapM ((liftM (\x -> H.dd $ (x >> nl opts))) . blockListToHtml opts) defs return $ mconcat $ nl opts : term' : nl opts : @@ -613,8 +606,22 @@ inlineToHtml opts inline = (Str str) -> return $ strToHtml str (Space) -> return $ strToHtml " " (LineBreak) -> return $ if writerHtml5 opts then H5.br else H.br - (Span attr ils) -> inlineListToHtml opts ils >>= - return . addAttrs opts attr . H.span + (Span (id',classes,kvs) ils) + -> inlineListToHtml opts ils >>= + return . addAttrs opts attr' . H.span + where attr' = (id',classes',kvs') + classes' = filter (`notElem` ["csl-no-emph", + "csl-no-strong", + "csl-no-smallcaps"]) classes + kvs' = if null styles + then kvs + else (("style", concat styles) : kvs) + styles = ["font-style:normal;" + | "csl-no-emph" `elem` classes] + ++ ["font-weight:normal;" + | "csl-no-strong" `elem` classes] + ++ ["font-variant:normal;" + | "csl-no-smallcaps" `elem` classes] (Emph lst) -> inlineListToHtml opts lst >>= return . H.em (Strong lst) -> inlineListToHtml opts lst >>= return . H.strong (Code attr str) -> case hlCode of @@ -742,7 +749,9 @@ inlineToHtml opts inline = else [A.title $ toValue tit]) return $ foldl (!) H5.embed attributes -- note: null title included, as in Markdown.pl - (Note contents) -> do + (Note contents) + | writerIgnoreNotes opts -> return mempty + | otherwise -> do st <- get let notes = stNotes st let number = (length notes) + 1 @@ -757,11 +766,11 @@ inlineToHtml opts inline = writerIdentifierPrefix opts ++ "fn" ++ ref) ! A.class_ "footnoteRef" ! prefixedId opts ("fnref" ++ ref) + $ H.sup $ toHtml ref - let link' = case writerEpubVersion opts of - Just EPUB3 -> link ! customAttribute "epub:type" "noteref" - _ -> link - return $ H.sup $ link' + return $ case writerEpubVersion opts of + Just EPUB3 -> link ! customAttribute "epub:type" "noteref" + _ -> link (Cite cits il)-> do contents <- inlineListToHtml opts il let citationIds = unwords $ map citationId cits let result = H.span ! A.class_ "citation" $ contents diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index dbb9b477a..b2eff4490 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -40,6 +40,7 @@ import Network.URI ( isURI, unEscapeString ) import Data.List ( (\\), isSuffixOf, isInfixOf, isPrefixOf, intercalate, intersperse ) import Data.Char ( toLower, isPunctuation, isAscii, isLetter, isDigit, ord ) +import Data.Maybe ( fromMaybe ) import Control.Applicative ((<|>)) import Control.Monad.State import Text.Pandoc.Pretty @@ -50,6 +51,8 @@ import Text.Pandoc.Highlighting (highlight, styleToLaTeX, data WriterState = WriterState { stInNote :: Bool -- true if we're in a note + , stInMinipage :: Bool -- true if in minipage + , stNotes :: [Doc] -- notes in a minipage , stOLLevel :: Int -- level of ordered list nesting , stOptions :: WriterOptions -- writer options, so they don't have to be parameter , stVerbInNote :: Bool -- true if document has verbatim text in note @@ -70,7 +73,7 @@ data WriterState = writeLaTeX :: WriterOptions -> Pandoc -> String writeLaTeX options document = evalState (pandocToLaTeX options document) $ - WriterState { stInNote = False, + WriterState { stInNote = False, stInMinipage = False, stNotes = [], stOLLevel = 1, stOptions = options, stVerbInNote = False, stTable = False, stStrikeout = False, @@ -126,14 +129,16 @@ pandocToLaTeX options (Pandoc meta blocks) = do (biblioTitle :: String) <- liftM (render colwidth) $ inlineListToLaTeX lastHeader let main = render colwidth $ vsep body st <- get + titleMeta <- stringToLaTeX TextString $ stringify $ docTitle meta + authorsMeta <- mapM (stringToLaTeX TextString . stringify) $ docAuthors meta let context = defField "toc" (writerTableOfContents options) $ defField "toc-depth" (show (writerTOCDepth options - if writerChapters options then 1 else 0)) $ defField "body" main $ - defField "title-meta" (stringify $ docTitle meta) $ - defField "author-meta" (intercalate "; " $ map stringify $ docAuthors meta) $ + defField "title-meta" titleMeta $ + defField "author-meta" (intercalate "; " authorsMeta) $ defField "documentclass" (if writerBeamer options then ("beamer" :: String) else if writerChapters options @@ -186,7 +191,7 @@ stringToLaTeX _ [] = return "" stringToLaTeX ctx (x:xs) = do opts <- gets stOptions rest <- stringToLaTeX ctx xs - let ligatures = writerTeXLigatures opts && not (ctx == CodeString) + let ligatures = writerTeXLigatures opts && (ctx /= CodeString) let isUrl = ctx == URLString when (x == '€') $ modify $ \st -> st{ stUsesEuro = True } @@ -236,7 +241,7 @@ inCmd cmd contents = char '\\' <> text cmd <> braces contents toSlides :: [Block] -> State WriterState [Block] toSlides bs = do opts <- gets stOptions - let slideLevel = maybe (getSlideLevel bs) id $ writerSlideLevel opts + let slideLevel = fromMaybe (getSlideLevel bs) $ writerSlideLevel opts let bs' = prepSlides slideLevel bs concat `fmap` (mapM (elementToBeamer slideLevel) $ hierarchicalize bs') @@ -439,7 +444,7 @@ blockToLaTeX (DefinitionList lst) = do incremental <- gets stIncremental let inc = if incremental then "[<+->]" else "" items <- mapM defListItemToLaTeX lst - let spacing = if and $ map isTightList (map snd lst) + let spacing = if all isTightList (map snd lst) then text "\\itemsep1pt\\parskip0pt\\parsep0pt" else empty return $ text ("\\begin{description}" ++ inc) $$ spacing $$ vcat items $$ @@ -451,12 +456,12 @@ blockToLaTeX (Header level (id',classes,_) lst) = blockToLaTeX (Table caption aligns widths heads rows) = do headers <- if all null heads then return empty - else ($$ "\\hline\\noalign{\\medskip}") `fmap` + else ($$ "\\midrule\\endhead") `fmap` (tableRowToLaTeX True aligns widths) heads captionText <- inlineListToLaTeX caption let capt = if isEmpty captionText then empty - else text "\\noalign{\\medskip}" + else text "\\addlinespace" $$ text "\\caption" <> braces captionText rows' <- mapM (tableRowToLaTeX False aligns widths) rows let colDescriptors = text $ concat $ map toColDescriptor aligns @@ -464,10 +469,10 @@ blockToLaTeX (Table caption aligns widths heads rows) = do return $ "\\begin{longtable}[c]" <> braces ("@{}" <> colDescriptors <> "@{}") -- the @{} removes extra space at beginning and end - $$ "\\hline\\noalign{\\medskip}" + $$ "\\toprule\\addlinespace" $$ headers $$ vcat rows' - $$ "\\hline" + $$ "\\bottomrule" $$ capt $$ "\\end{longtable}" @@ -488,23 +493,42 @@ tableRowToLaTeX :: Bool -> [[Block]] -> State WriterState Doc tableRowToLaTeX header aligns widths cols = do - renderedCells <- mapM blockListToLaTeX cols - let valign = text $ if header then "[b]" else "[t]" - let halign x = case x of - AlignLeft -> "\\raggedright" - AlignRight -> "\\raggedleft" - AlignCenter -> "\\centering" - AlignDefault -> "\\raggedright" -- scale factor compensates for extra space between columns -- so the whole table isn't larger than columnwidth let scaleFactor = 0.97 ** fromIntegral (length aligns) - let toCell 0 _ c = c - toCell w a c = "\\begin{minipage}" <> valign <> - braces (text (printf "%.2f\\columnwidth" - (w * scaleFactor))) <> - (halign a <> cr <> c <> cr) <> "\\end{minipage}" - let cells = zipWith3 toCell widths aligns renderedCells - return $ hsep (intersperse "&" cells) $$ "\\\\\\noalign{\\medskip}" + let widths' = map (scaleFactor *) widths + cells <- mapM (tableCellToLaTeX header) $ zip3 widths' aligns cols + return $ hsep (intersperse "&" cells) $$ "\\\\\\addlinespace" + +tableCellToLaTeX :: Bool -> (Double, Alignment, [Block]) + -> State WriterState Doc +tableCellToLaTeX _ (0, _, blocks) = blockListToLaTeX blocks +tableCellToLaTeX header (width, align, blocks) = do + modify $ \st -> st{ stInMinipage = True, stNotes = [] } + cellContents <- blockListToLaTeX blocks + notes <- gets stNotes + modify $ \st -> st{ stInMinipage = False, stNotes = [] } + let valign = text $ if header then "[b]" else "[t]" + let halign = case align of + AlignLeft -> "\\raggedright" + AlignRight -> "\\raggedleft" + AlignCenter -> "\\centering" + AlignDefault -> "\\raggedright" + return $ ("\\begin{minipage}" <> valign <> + braces (text (printf "%.2f\\columnwidth" width)) <> + (halign <> cr <> cellContents <> cr) <> "\\end{minipage}") + $$ case notes of + [] -> empty + ns -> (case length ns of + n | n > 1 -> "\\addtocounter" <> + braces "footnote" <> + braces (text $ show $ 1 - n) + | otherwise -> empty) + $$ + vcat (intersperse + ("\\addtocounter" <> braces "footnote" <> braces "1") + $ map (\x -> "\\footnotetext" <> braces x) + $ reverse ns) listItemToLaTeX :: [Block] -> State WriterState Doc listItemToLaTeX lst = blockListToLaTeX lst >>= return . (text "\\item" $$) . @@ -600,7 +624,16 @@ isQuoted _ = False -- | Convert inline element to LaTeX inlineToLaTeX :: Inline -- ^ Inline to convert -> State WriterState Doc -inlineToLaTeX (Span _ ils) = inlineListToLaTeX ils >>= return . braces +inlineToLaTeX (Span (_,classes,_) ils) = do + let noEmph = "csl-no-emph" `elem` classes + let noStrong = "csl-no-strong" `elem` classes + let noSmallCaps = "csl-no-smallcaps" `elem` classes + ((if noEmph then inCmd "textup" else id) . + (if noStrong then inCmd "textnormal" else id) . + (if noSmallCaps then inCmd "textnormal" else id) . + (if not (noEmph || noStrong || noSmallCaps) + then braces + else id)) `fmap` inlineListToLaTeX ils inlineToLaTeX (Emph lst) = inlineListToLaTeX lst >>= return . inCmd "emph" inlineToLaTeX (Strong lst) = @@ -698,14 +731,20 @@ inlineToLaTeX (Image _ (source, _)) = do source'' <- stringToLaTeX URLString source' return $ "\\includegraphics" <> braces (text source'') inlineToLaTeX (Note contents) = do + inMinipage <- gets stInMinipage modify (\s -> s{stInNote = True}) contents' <- blockListToLaTeX contents modify (\s -> s {stInNote = False}) let optnl = case reverse contents of (CodeBlock _ _ : _) -> cr _ -> empty - return $ "\\footnote" <> braces (nest 2 contents' <> optnl) - -- note: a \n before } needed when note ends with a Verbatim environment + let noteContents = nest 2 contents' <> optnl + modify $ \st -> st{ stNotes = noteContents : stNotes st } + return $ + if inMinipage + then "\\footnotemark{}" + -- note: a \n before } needed when note ends with a Verbatim environment + else "\\footnote" <> braces noteContents citationsToNatbib :: [Citation] -> State WriterState Doc citationsToNatbib (one:[]) @@ -726,9 +765,9 @@ citationsToNatbib cits | noPrefix (tail cits) && noSuffix (init cits) && ismode NormalCitation cits = citeCommand "citep" p s ks where - noPrefix = and . map (null . citationPrefix) - noSuffix = and . map (null . citationSuffix) - ismode m = and . map (((==) m) . citationMode) + noPrefix = all (null . citationPrefix) + noSuffix = all (null . citationSuffix) + ismode m = all (((==) m) . citationMode) p = citationPrefix $ head $ cits s = citationSuffix $ last $ cits ks = intercalate ", " $ map citationId cits diff --git a/src/Text/Pandoc/Writers/Markdown.hs b/src/Text/Pandoc/Writers/Markdown.hs index eefcd547a..278e5cc9d 100644 --- a/src/Text/Pandoc/Writers/Markdown.hs +++ b/src/Text/Pandoc/Writers/Markdown.hs @@ -338,7 +338,7 @@ blockToMarkdown opts (RawBlock f str) else return $ if isEnabled Ext_markdown_attribute opts then text (addMarkdownAttribute str) <> text "\n" else text str <> text "\n" - | f == "latex" || f == "tex" || f == "markdown" = do + | f `elem` ["latex", "tex", "markdown"] = do st <- get if stPlain st then return empty @@ -381,13 +381,11 @@ blockToMarkdown opts (CodeBlock (_,classes,_) str) isEnabled Ext_literate_haskell opts = return $ prefixed "> " (text str) <> blankline blockToMarkdown opts (CodeBlock attribs str) = return $ - case attribs of - x | x /= nullAttr && isEnabled Ext_fenced_code_blocks opts -> - tildes <> " " <> attrs <> cr <> text str <> - cr <> tildes <> blankline - (_,(cls:_),_) | isEnabled Ext_backtick_code_blocks opts -> - backticks <> " " <> text cls <> cr <> text str <> - cr <> backticks <> blankline + case attribs == nullAttr of + False | isEnabled Ext_backtick_code_blocks opts -> + backticks <> attrs <> cr <> text str <> cr <> backticks <> blankline + | isEnabled Ext_fenced_code_blocks opts -> + tildes <> attrs <> cr <> text str <> cr <> tildes <> blankline _ -> nest (writerTabStop opts) (text str) <> blankline where tildes = text $ case [ln | ln <- lines str, all (=='~') ln] of [] -> "~~~~" @@ -396,8 +394,10 @@ blockToMarkdown opts (CodeBlock attribs str) = return $ | otherwise -> replicate (n+1) '~' backticks = text "```" attrs = if isEnabled Ext_fenced_code_attributes opts - then nowrap $ attrsToMarkdown attribs - else empty + then nowrap $ " " <> attrsToMarkdown attribs + else case attribs of + (_,[cls],_) -> " " <> text cls + _ -> empty blockToMarkdown opts (BlockQuote blocks) = do st <- get -- if we're writing literate haskell, put a space before the bird tracks @@ -555,7 +555,14 @@ bulletListItemToMarkdown opts items = do contents <- blockListToMarkdown opts items let sps = replicate (writerTabStop opts - 2) ' ' let start = text ('-' : ' ' : sps) - return $ hang (writerTabStop opts) start $ contents <> cr + -- remove trailing blank line if it is a tight list + let contents' = case reverse items of + (BulletList xs:_) | isTightList xs -> + chomp contents <> cr + (OrderedList _ xs:_) | isTightList xs -> + chomp contents <> cr + _ -> contents + return $ hang (writerTabStop opts) start $ contents' <> cr -- | Convert ordered list item (a list of blocks) to markdown. orderedListItemToMarkdown :: WriterOptions -- ^ options @@ -621,10 +628,11 @@ getReference label (src, tit) = do Nothing -> do let label' = case find ((== label) . fst) (stRefs st) of Just _ -> -- label is used; generate numerical label - case find (\n -> not (any (== [Str (show n)]) - (map fst (stRefs st)))) [1..(10000 :: Integer)] of - Just x -> [Str (show x)] - Nothing -> error "no unique label" + case find (\n -> notElem [Str (show n)] + (map fst (stRefs st))) + [1..(10000 :: Integer)] of + Just x -> [Str (show x)] + Nothing -> error "no unique label" Nothing -> label modify (\s -> s{ stRefs = (label', (src,tit)) : stRefs st }) return label' diff --git a/src/Text/Pandoc/Writers/MediaWiki.hs b/src/Text/Pandoc/Writers/MediaWiki.hs index 61741a61e..83fefaa29 100644 --- a/src/Text/Pandoc/Writers/MediaWiki.hs +++ b/src/Text/Pandoc/Writers/MediaWiki.hs @@ -51,7 +51,7 @@ data WriterState = WriterState { writeMediaWiki :: WriterOptions -> Pandoc -> String writeMediaWiki opts document = evalState (pandocToMediaWiki opts document) - (WriterState { stNotes = False, stListLevel = [], stUseTags = False }) + WriterState { stNotes = False, stListLevel = [], stUseTags = False } -- | Return MediaWiki representation of document. pandocToMediaWiki :: WriterOptions -> Pandoc -> State WriterState String diff --git a/src/Text/Pandoc/Writers/ODT.hs b/src/Text/Pandoc/Writers/ODT.hs index cc0a06243..c3652d65d 100644 --- a/src/Text/Pandoc/Writers/ODT.hs +++ b/src/Text/Pandoc/Writers/ODT.hs @@ -30,7 +30,10 @@ Conversion of 'Pandoc' documents to ODT. -} module Text.Pandoc.Writers.ODT ( writeODT ) where import Data.IORef -import Data.List ( isPrefixOf ) +import Data.List ( isPrefixOf, isSuffixOf ) +import Data.Maybe ( fromMaybe ) +import Text.XML.Light.Output +import Text.TeXMath import qualified Data.ByteString.Lazy as B import Text.Pandoc.UTF8 ( fromStringLazy ) import Codec.Archive.Zip @@ -40,13 +43,14 @@ import Text.Pandoc.ImageSize ( imageSize, sizeInPoints ) import Text.Pandoc.MIME ( getMimeType ) import Text.Pandoc.Definition import Text.Pandoc.Walk +import Text.Pandoc.Writers.Shared ( fixDisplayMath ) import Text.Pandoc.Writers.OpenDocument ( writeOpenDocument ) import Control.Monad (liftM) import Text.Pandoc.XML import Text.Pandoc.Pretty import qualified Control.Exception as E import Data.Time.Clock.POSIX ( getPOSIXTime ) -import System.FilePath ( takeExtension ) +import System.FilePath ( takeExtension, takeDirectory ) -- | Produce an ODT file from a Pandoc document. writeODT :: WriterOptions -- ^ Writer options @@ -60,9 +64,9 @@ writeODT opts doc@(Pandoc meta _) = do Just f -> B.readFile f Nothing -> (B.fromChunks . (:[])) `fmap` readDataFile datadir "reference.odt" - -- handle pictures + -- handle formulas and pictures picEntriesRef <- newIORef ([] :: [Entry]) - doc' <- walkM (transformPic opts picEntriesRef) doc + doc' <- walkM (transformPicMath opts picEntriesRef) $ walk fixDisplayMath doc let newContents = writeOpenDocument opts{writerWrapText = False} doc' epochtime <- floor `fmap` getPOSIXTime let contentEntry = toEntry "content.xml" epochtime @@ -72,7 +76,11 @@ writeODT opts doc@(Pandoc meta _) = do $ contentEntry : picEntries -- construct META-INF/manifest.xml based on archive let toFileEntry fp = case getMimeType fp of - Nothing -> empty + Nothing -> if "Formula-" `isPrefixOf` fp && "/" `isSuffixOf` fp + then selfClosingTag "manifest:file-entry" + [("manifest:media-type","application/vnd.oasis.opendocument.formula") + ,("manifest:full-path",fp)] + else empty Just m -> selfClosingTag "manifest:file-entry" [("manifest:media-type", m) ,("manifest:full-path", fp) @@ -80,6 +88,8 @@ writeODT opts doc@(Pandoc meta _) = do ] let files = [ ent | ent <- filesInArchive archive, not ("META-INF" `isPrefixOf` ent) ] + let formulas = [ takeDirectory ent ++ "/" | ent <- filesInArchive archive, + "Formula-" `isPrefixOf` ent, takeExtension ent == ".xml" ] let manifestEntry = toEntry "META-INF/manifest.xml" epochtime $ fromStringLazy $ render Nothing $ text "<?xml version=\"1.0\" encoding=\"utf-8\"?>" @@ -91,6 +101,7 @@ writeODT opts doc@(Pandoc meta _) = do [("manifest:media-type","application/vnd.oasis.opendocument.text") ,("manifest:full-path","/")] $$ vcat ( map toFileEntry $ files ) + $$ vcat ( map toFileEntry $ formulas ) ) ) let archive' = addEntryToArchive manifestEntry archive @@ -118,8 +129,8 @@ writeODT opts doc@(Pandoc meta _) = do $ addEntryToArchive metaEntry archive' return $ fromArchive archive'' -transformPic :: WriterOptions -> IORef [Entry] -> Inline -> IO Inline -transformPic opts entriesRef (Image lab (src,_)) = do +transformPicMath :: WriterOptions -> IORef [Entry] -> Inline -> IO Inline +transformPicMath opts entriesRef (Image lab (src,_)) = do res <- fetchItem (writerSourceURL opts) src case res of Left (_ :: E.SomeException) -> do @@ -127,7 +138,7 @@ transformPic opts entriesRef (Image lab (src,_)) = do return $ Emph lab Right (img, _) -> do let size = imageSize img - let (w,h) = maybe (0,0) id $ sizeInPoints `fmap` size + let (w,h) = fromMaybe (0,0) $ sizeInPoints `fmap` size let tit' = show w ++ "x" ++ show h entries <- readIORef entriesRef let newsrc = "Pictures/" ++ show (length entries) ++ takeExtension src @@ -136,5 +147,29 @@ transformPic opts entriesRef (Image lab (src,_)) = do let entry = toEntry newsrc epochtime $ toLazy img modifyIORef entriesRef (entry:) return $ Image lab (newsrc, tit') -transformPic _ _ x = return x +transformPicMath _ entriesRef (Math t math) = do + entries <- readIORef entriesRef + let dt = if t == InlineMath then DisplayInline else DisplayBlock + case texMathToMathML dt math of + Left _ -> return $ Math t math + Right r -> do + let conf = useShortEmptyTags (const False) defaultConfigPP + let mathml = ppcTopElement conf r + epochtime <- floor `fmap` getPOSIXTime + let dirname = "Formula-" ++ show (length entries) ++ "/" + let fname = dirname ++ "content.xml" + let entry = toEntry fname epochtime (fromStringLazy mathml) + modifyIORef entriesRef (entry:) + return $ RawInline (Format "opendocument") $ render Nothing $ + inTags False "draw:frame" [("text:anchor-type", + if t == DisplayMath + then "paragraph" + else "as-char") + ,("style:vertical-pos", "middle") + ,("style:vertical-rel", "text")] $ + selfClosingTag "draw:object" [("xlink:href", dirname) + , ("xlink:type", "simple") + , ("xlink:show", "embed") + , ("xlink:actuate", "onLoad")] +transformPicMath _ _ x = return x diff --git a/src/Text/Pandoc/Writers/OpenDocument.hs b/src/Text/Pandoc/Writers/OpenDocument.hs index b38d250aa..0029c3296 100644 --- a/src/Text/Pandoc/Writers/OpenDocument.hs +++ b/src/Text/Pandoc/Writers/OpenDocument.hs @@ -64,6 +64,7 @@ data WriterState = , stInDefinition :: Bool , stTight :: Bool , stFirstPara :: Bool + , stImageId :: Int } defaultWriterState :: WriterState @@ -78,6 +79,7 @@ defaultWriterState = , stInDefinition = False , stTight = False , stFirstPara = False + , stImageId = 1 } when :: Bool -> Doc -> Doc @@ -283,8 +285,12 @@ blocksToOpenDocument o b = vcat <$> mapM (blockToOpenDocument o) b -- | Convert a Pandoc block element to OpenDocument. blockToOpenDocument :: WriterOptions -> Block -> State WriterState Doc blockToOpenDocument o bs - | Plain b <- bs = inParagraphTags =<< inlinesToOpenDocument o b - | Para b <- bs = inParagraphTags =<< inlinesToOpenDocument o b + | Plain b <- bs = if null b + then return empty + else inParagraphTags =<< inlinesToOpenDocument o b + | Para b <- bs = if null b + then return empty + else inParagraphTags =<< inlinesToOpenDocument o b | Div _ xs <- bs = blocksToOpenDocument o xs | Header i _ b <- bs = setFirstPara >> (inHeaderTags i =<< inlinesToOpenDocument o b) @@ -296,8 +302,8 @@ blockToOpenDocument o bs | Table c a w h r <- bs = setFirstPara >> table c a w h r | HorizontalRule <- bs = setFirstPara >> return (selfClosingTag "text:p" [ ("text:style-name", "Horizontal_20_Line") ]) - | RawBlock f s <- bs = if f == "opendocument" - then preformatted s + | RawBlock f s <- bs = if f == Format "opendocument" + then return $ text s else return empty | Null <- bs = return empty | otherwise = return empty @@ -376,11 +382,11 @@ inlineToOpenDocument o ils | Code _ s <- ils = withTextStyle Pre $ inTextStyle $ preformatted s | Math t s <- ils = inlinesToOpenDocument o (readTeXMath' t s) | Cite _ l <- ils = inlinesToOpenDocument o l - | RawInline f s <- ils = if f == "opendocument" || f == "html" - then withTextStyle Pre $ inTextStyle $ preformatted s + | RawInline f s <- ils = if f == Format "opendocument" + then return $ text s else return empty | Link l (s,t) <- ils = mkLink s t <$> inlinesToOpenDocument o l - | Image _ (s,t) <- ils = return $ mkImg s t + | Image _ (s,t) <- ils = mkImg s t | Note l <- ils = mkNote l | otherwise = return empty where @@ -389,7 +395,11 @@ inlineToOpenDocument o ils , ("xlink:href" , s ) , ("office:name", t ) ] . inSpanTags "Definition" - mkImg s t = inTags False "draw:frame" (attrsFromTitle t) $ + mkImg s t = do + id' <- gets stImageId + modify (\st -> st{ stImageId = id' + 1 }) + return $ inTags False "draw:frame" + (("draw:name", "img" ++ show id'):attrsFromTitle t) $ selfClosingTag "draw:image" [ ("xlink:href" , s ) , ("xlink:type" , "simple") , ("xlink:show" , "embed" ) diff --git a/src/Text/Pandoc/Writers/Org.hs b/src/Text/Pandoc/Writers/Org.hs index 51083f52b..d318c5f6a 100644 --- a/src/Text/Pandoc/Writers/Org.hs +++ b/src/Text/Pandoc/Writers/Org.hs @@ -129,7 +129,7 @@ blockToOrg (Para inlines) = do blockToOrg (RawBlock "html" str) = return $ blankline $$ "#+BEGIN_HTML" $$ nest 2 (text str) $$ "#+END_HTML" $$ blankline -blockToOrg (RawBlock f str) | f == "org" || f == "latex" || f == "tex" = +blockToOrg (RawBlock f str) | f `elem` ["org", "latex", "tex"] = return $ text str blockToOrg (RawBlock _ _) = return empty blockToOrg HorizontalRule = return $ blankline $$ "--------------" $$ blankline diff --git a/src/Text/Pandoc/Writers/RST.hs b/src/Text/Pandoc/Writers/RST.hs index 1a62f7250..37bb66632 100644 --- a/src/Text/Pandoc/Writers/RST.hs +++ b/src/Text/Pandoc/Writers/RST.hs @@ -287,7 +287,7 @@ definitionListItemToRST (label, defs) = do label' <- inlineListToRST label contents <- liftM vcat $ mapM blockListToRST defs tabstop <- get >>= (return . writerTabStop . stOptions) - return $ label' $$ nest tabstop (contents <> cr) + return $ label' $$ nest tabstop (nestle contents <> cr) -- | Convert list of Pandoc block elements to RST. blockListToRST :: [Block] -- ^ List of block elements diff --git a/src/Text/Pandoc/Writers/Shared.hs b/src/Text/Pandoc/Writers/Shared.hs index 9cb08803c..604aac1c9 100644 --- a/src/Text/Pandoc/Writers/Shared.hs +++ b/src/Text/Pandoc/Writers/Shared.hs @@ -34,6 +34,7 @@ module Text.Pandoc.Writers.Shared ( , setField , defField , tagWithAttrs + , fixDisplayMath ) where import Text.Pandoc.Definition @@ -46,6 +47,7 @@ import qualified Data.Map as M import qualified Data.Text as T import Data.Aeson (FromJSON(..), fromJSON, ToJSON (..), Value(Object), Result(..)) import qualified Data.Traversable as Traversable +import Data.List ( groupBy ) -- | Create JSON value for template from a 'Meta' and an association list -- of variables, specified at the command line or in the writer. @@ -65,8 +67,7 @@ metaToJSON opts blockWriter inlineWriter (Meta metamap) renderedMap <- Traversable.mapM (metaValueToJSON blockWriter inlineWriter) metamap - return $ M.foldWithKey (\key val obj -> defField key val obj) - baseContext renderedMap + return $ M.foldWithKey defField baseContext renderedMap | otherwise = return (Object H.empty) metaValueToJSON :: Monad m @@ -137,3 +138,28 @@ tagWithAttrs tag (ident,classes,kvs) = hsep ,hsep (map (\(k,v) -> text k <> "=" <> doubleQuotes (text (escapeStringForXML v))) kvs) ] <> ">" + +isDisplayMath :: Inline -> Bool +isDisplayMath (Math DisplayMath _) = True +isDisplayMath _ = False + +stripLeadingTrailingSpace :: [Inline] -> [Inline] +stripLeadingTrailingSpace = go . reverse . go . reverse + where go (Space:xs) = xs + go xs = xs + +-- Put display math in its own block (for ODT/DOCX). +fixDisplayMath :: Block -> Block +fixDisplayMath (Plain lst) + | any isDisplayMath lst && not (all isDisplayMath lst) = + -- chop into several paragraphs so each displaymath is its own + Div ("",["math"],[]) $ map (Plain . stripLeadingTrailingSpace) $ + groupBy (\x y -> (isDisplayMath x && isDisplayMath y) || + not (isDisplayMath x || isDisplayMath y)) lst +fixDisplayMath (Para lst) + | any isDisplayMath lst && not (all isDisplayMath lst) = + -- chop into several paragraphs so each displaymath is its own + Div ("",["math"],[]) $ map (Para . stripLeadingTrailingSpace) $ + groupBy (\x y -> (isDisplayMath x && isDisplayMath y) || + not (isDisplayMath x || isDisplayMath y)) lst +fixDisplayMath x = x diff --git a/src/Text/Pandoc/Writers/Texinfo.hs b/src/Text/Pandoc/Writers/Texinfo.hs index d62e50880..bf3df8035 100644 --- a/src/Text/Pandoc/Writers/Texinfo.hs +++ b/src/Text/Pandoc/Writers/Texinfo.hs @@ -293,7 +293,7 @@ blockListToTexinfo (x:xs) = do case x of Header level _ _ -> do -- We need need to insert a menu for this node. - let (before, after) = break isHeader xs + let (before, after) = break isHeaderBlock xs before' <- blockListToTexinfo before let menu = if level < 4 then collectNodes (level + 1) after @@ -315,10 +315,6 @@ blockListToTexinfo (x:xs) = do xs' <- blockListToTexinfo xs return $ x' $$ xs' -isHeader :: Block -> Bool -isHeader (Header _ _ _) = True -isHeader _ = False - collectNodes :: Int -> [Block] -> [Block] collectNodes _ [] = [] collectNodes level (x:xs) = diff --git a/src/Text/Pandoc/Writers/Textile.hs b/src/Text/Pandoc/Writers/Textile.hs index 7c102cc86..95aedf780 100644 --- a/src/Text/Pandoc/Writers/Textile.hs +++ b/src/Text/Pandoc/Writers/Textile.hs @@ -51,7 +51,7 @@ data WriterState = WriterState { writeTextile :: WriterOptions -> Pandoc -> String writeTextile opts document = evalState (pandocToTextile opts document) - (WriterState { stNotes = [], stListLevel = [], stUseTags = False }) + WriterState { stNotes = [], stListLevel = [], stUseTags = False } -- | Return Textile representation of document. pandocToTextile :: WriterOptions -> Pandoc -> State WriterState String |