From 56f56e5e1594ef5d18326d1eb6de3176db307c6a Mon Sep 17 00:00:00 2001 From: Merijn Verstraaten Date: Sat, 7 Sep 2013 18:58:16 +0100 Subject: Added support for LaTeX style literate Haskell code blocks in rST. --- src/Text/Pandoc/Readers/RST.hs | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/RST.hs b/src/Text/Pandoc/Readers/RST.hs index 32893128a..c12a1493a 100644 --- a/src/Text/Pandoc/Readers/RST.hs +++ b/src/Text/Pandoc/Readers/RST.hs @@ -347,14 +347,25 @@ lhsCodeBlock = try $ do getPosition >>= guard . (==1) . sourceColumn guardEnabled Ext_literate_haskell optional codeBlockStart - lns <- many1 birdTrackLine - -- if (as is normal) there is always a space after >, drop it - let lns' = if all (\ln -> null ln || take 1 ln == " ") lns - then map (drop 1) lns - else lns + lns <- latexCodeBlock <|> birdCodeBlock blanklines return $ B.codeBlockWith ("", ["sourceCode", "literate", "haskell"], []) - $ intercalate "\n" lns' + $ intercalate "\n" lns + +latexCodeBlock :: Parser [Char] st [[Char]] +latexCodeBlock = try $ do + try (latexBlockLine "\\begin{code}") + many1Till anyLine (try $ latexBlockLine "\\end{code}") + where + latexBlockLine s = skipMany spaceChar >> string s >> blankline + +birdCodeBlock :: Parser [Char] st [[Char]] +birdCodeBlock = filterSpace <$> many1 birdTrackLine + where filterSpace lns = + -- if (as is normal) there is always a space after >, drop it + if all (\ln -> null ln || take 1 ln == " ") lns + then map (drop 1) lns + else lns birdTrackLine :: Parser [Char] st [Char] birdTrackLine = char '>' >> anyLine -- cgit v1.2.3 From 5b99112f229db1fcbe82943678f9f55c1ead8f11 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Wed, 6 Nov 2013 19:18:24 -0800 Subject: Docx writer: Fix URL for core-properties in `_rels/.rels`. Partially addresses #1046. --- src/Text/Pandoc/Writers/Docx.hs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 0fdea0a7a..f1276c831 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -247,7 +247,7 @@ 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 -- cgit v1.2.3 From 2efd0951d3d560a159f92df4730e2cee26978698 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 7 Nov 2013 08:46:52 -0800 Subject: Docx writer: fixed core metadata. - Don't create empty date nodes if no date given. - Don't create multiple dc:creator nodes; instead separate by semicolons. Closes #1046. --- src/Text/Pandoc/Writers/Docx.hs | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index f1276c831..5d1647844 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -231,10 +231,11 @@ 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" -- cgit v1.2.3 From 01fed75b8f755bf4de3d4ec2b720e4dd9e66951a Mon Sep 17 00:00:00 2001 From: MinRK Date: Thu, 7 Nov 2013 22:25:44 -0800 Subject: recognize svg tag in HTML Reader avoids adding lots of `

` tags in embedded SVG content, for instance in markdown to HTML. --- src/Text/Pandoc/Readers/HTML.hs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index 80279bf61..d691c9878 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -557,7 +557,7 @@ blockHtmlTags = ["address", "article", "aside", "blockquote", "body", "button", "noframes", "noscript", "object", "ol", "output", "p", "pre", "progress", "section", "table", "tbody", "textarea", "thead", "tfoot", "ul", "dd", "dt", "frameset", "li", "tbody", "td", "tfoot", - "th", "thead", "tr", "script", "style", "video"] + "th", "thead", "tr", "script", "style", "svg", "video"] -- We want to allow raw docbook in markdown documents, so we -- include docbook block tags here too. -- cgit v1.2.3 From b4441c940dc76d1a64636ef88287b38306ebccb4 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 12 Nov 2013 18:48:06 -0800 Subject: HTML/EPUB footnotes: Put `` tag inside `` tags. This allows better control of formatting, since the `` tags have a distinguishing class. Closes #1049. --- src/Text/Pandoc/Writers/HTML.hs | 8 ++++---- tests/writer.html | 6 +++--- 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index c1cca291b..424843539 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -757,11 +757,11 @@ 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/tests/writer.html b/tests/writer.html index e0d1a3b25..d00b8ca66 100644 --- a/tests/writer.html +++ b/tests/writer.html @@ -544,12 +544,12 @@ document.write(''+e+'<\/'+'a'+'>');

Here is a movie movie icon.

Footnotes

-

Here is a footnote reference,1 and another.2 This should not be a footnote reference, because it contains a space.[^my note] Here is an inline note.3

+

Here is a footnote reference,1 and another.2 This should not be a footnote reference, because it contains a space.[^my note] Here is an inline note.3

-

Notes can go in quotes.4

+

Notes can go in quotes.4

    -
  1. And in list items.5
    2. +
  2. And in list items.5

This paragraph should not be part of the note, as it is not indented.

-- cgit v1.2.3 From 892ba2dd163cbba598cbd86814a594008aae9f27 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 16 Nov 2013 14:09:09 -0800 Subject: LaTeX writer: Properly escape pdftitle, pdfauthor. Closes #1059. --- src/Text/Pandoc/Writers/LaTeX.hs | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index 72b0bde6d..7ff64cf74 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -126,14 +126,16 @@ 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 -- cgit v1.2.3 From 9b0378b939f46183fc152b1a49a69f3007de295a Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 16 Nov 2013 22:49:15 -0800 Subject: OpenDocument: Skip raw HTML. Previously it was erroneously included as verbatim text. Closes #1035. --- src/Text/Pandoc/Writers/OpenDocument.hs | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/OpenDocument.hs b/src/Text/Pandoc/Writers/OpenDocument.hs index b38d250aa..565f5f869 100644 --- a/src/Text/Pandoc/Writers/OpenDocument.hs +++ b/src/Text/Pandoc/Writers/OpenDocument.hs @@ -376,8 +376,8 @@ 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 == "opendocument" + then return $ preformatted s else return empty | Link l (s,t) <- ils = mkLink s t <$> inlinesToOpenDocument o l | Image _ (s,t) <- ils = return $ mkImg s t -- cgit v1.2.3 From d5660275a38a58334372326a79a9ce0153fede43 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 17 Nov 2013 08:45:21 -0800 Subject: Parsing: Generalized type of registerHeader, using new typeclasses. New type classes HasReadeOptions, HasIdentifierList, HasHeaderMap. These allow certain common functions to be reused even in parsers that use custom state (instead of ParserState), such as the MediaWiki reader. Minor API bump. --- src/Text/Pandoc/Parsing.hs | 54 +++++++++++++++++++++++++++++++++++----------- 1 file changed, 42 insertions(+), 12 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Parsing.hs b/src/Text/Pandoc/Parsing.hs index 701b2ef84..9687d7712 100644 --- a/src/Text/Pandoc/Parsing.hs +++ b/src/Text/Pandoc/Parsing.hs @@ -1,4 +1,5 @@ -{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE GeneralizedNewtypeDeriving, TypeSynonymInstances, + FlexibleInstances#-} {- Copyright (C) 2006-2010 John MacFarlane @@ -65,6 +66,9 @@ module Text.Pandoc.Parsing ( (>>~), guardEnabled, guardDisabled, ParserState (..), + HasReaderOptions (..), + HasHeaderMap (..), + HasIdentifierList (..), defaultParserState, HeaderType (..), ParserContext (..), @@ -826,6 +830,34 @@ instance HasMeta ParserState where deleteMeta field st = st{ stateMeta = deleteMeta field $ stateMeta st } +class Monad m => HasReaderOptions m where + askReaderOption :: (ReaderOptions -> b) -> m b + +class Monad m => HasHeaderMap m where + getHeaderMap :: m (M.Map Inlines String) + putHeaderMap :: M.Map Inlines String -> m () + modifyHeaderMap :: (M.Map Inlines String -> M.Map Inlines String) -> m () + -- default + modifyHeaderMap f = getHeaderMap >>= putHeaderMap . f + +class Monad m => HasIdentifierList m where + getIdentifierList :: m [String] + putIdentifierList :: [String] -> m () + modifyIdentifierList :: ([String] -> [String]) -> m () + -- default + modifyIdentifierList f = getIdentifierList >>= putIdentifierList . f + +instance HasReaderOptions (Parser s ParserState) where + askReaderOption = getOption + +instance HasHeaderMap (Parser s ParserState) where + getHeaderMap = fmap stateHeaders getState + putHeaderMap hm = updateState $ \st -> st{ stateHeaders = hm } + +instance HasIdentifierList (Parser s ParserState) where + getIdentifierList = fmap stateIdentifiers getState + putIdentifierList l = updateState $ \st -> st{ stateIdentifiers = l } + defaultParserState :: ParserState defaultParserState = ParserState { stateOptions = def, @@ -895,10 +927,11 @@ type SubstTable = M.Map Key Inlines -- and the auto_identifers extension is set, generate a new -- unique identifier, and update the list of identifiers -- in state. -registerHeader :: Attr -> Inlines -> Parser s ParserState Attr +registerHeader :: (HasReaderOptions m, HasHeaderMap m, HasIdentifierList m) + => Attr -> Inlines -> m Attr registerHeader (ident,classes,kvs) header' = do - ids <- stateIdentifiers `fmap` getState - exts <- getOption readerExtensions + ids <- getIdentifierList + exts <- askReaderOption readerExtensions let insert' = M.insertWith (\_new old -> old) if null ident && Ext_auto_identifiers `Set.member` exts then do @@ -906,16 +939,13 @@ registerHeader (ident,classes,kvs) header' = do let id'' = if Ext_ascii_identifiers `Set.member` exts then catMaybes $ map toAsciiChar id' else id' - updateState $ \st -> st{ - stateIdentifiers = if id' == id'' - then id' : ids - else id' : id'' : ids, - stateHeaders = insert' header' id' $ stateHeaders st } + putIdentifierList $ if id' == id'' + then id' : ids + else id' : id'' : ids + modifyHeaderMap $ insert' header' id' return (id'',classes,kvs) else do - unless (null ident) $ - updateState $ \st -> st{ - stateHeaders = insert' header' ident $ stateHeaders st } + unless (null ident) $ modifyHeaderMap $ insert' header' ident return (ident,classes,kvs) -- | Fail unless we're in "smart typography" mode. -- cgit v1.2.3 From 0fd2176e29bd1118d314c6179455fb78bed35aea Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 17 Nov 2013 08:47:14 -0800 Subject: MediaWiki reader: Add automatic header identifiers. --- src/Text/Pandoc/Readers/MediaWiki.hs | 22 ++++++++++++-- tests/mediawiki-reader.native | 58 ++++++++++++++++++------------------ 2 files changed, 49 insertions(+), 31 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/MediaWiki.hs b/src/Text/Pandoc/Readers/MediaWiki.hs index 136701bd0..1c074e3de 100644 --- a/src/Text/Pandoc/Readers/MediaWiki.hs +++ b/src/Text/Pandoc/Readers/MediaWiki.hs @@ -1,4 +1,5 @@ -{-# LANGUAGE RelaxedPolyRec #-} -- needed for inlinesBetween on GHC < 7 +{-# LANGUAGE RelaxedPolyRec, FlexibleInstances, TypeSynonymInstances #-} +-- RelaxedPolyRec needed for inlinesBetween on GHC < 7 {- Copyright (C) 2012 John MacFarlane @@ -51,6 +52,7 @@ import Data.List (intersperse, intercalate, isPrefixOf ) import Text.HTML.TagSoup import Data.Sequence (viewl, ViewL(..), (<|)) import qualified Data.Foldable as F +import qualified Data.Map as M import Data.Char (isDigit, isSpace) -- | Read mediawiki from an input string and return a Pandoc document. @@ -62,6 +64,8 @@ readMediaWiki opts s = , mwMaxNestingLevel = 4 , mwNextLinkNumber = 1 , mwCategoryLinks = [] + , mwHeaderMap = M.empty + , mwIdentifierList = [] } "source" (s ++ "\n") of Left err' -> error $ "\nError:\n" ++ show err' @@ -71,10 +75,23 @@ data MWState = MWState { mwOptions :: ReaderOptions , mwMaxNestingLevel :: Int , mwNextLinkNumber :: Int , mwCategoryLinks :: [Inlines] + , mwHeaderMap :: M.Map Inlines String + , mwIdentifierList :: [String] } type MWParser = Parser [Char] MWState +instance HasReaderOptions MWParser where + askReaderOption f = (f . mwOptions) `fmap` getState + +instance HasHeaderMap MWParser where + getHeaderMap = fmap mwHeaderMap getState + putHeaderMap hm = updateState $ \st -> st{ mwHeaderMap = hm } + +instance HasIdentifierList MWParser where + getIdentifierList = fmap mwIdentifierList getState + putIdentifierList l = updateState $ \st -> st{ mwIdentifierList = l } + -- -- auxiliary functions -- @@ -351,7 +368,8 @@ header = try $ do let lev = length eqs guard $ lev <= 6 contents <- trimInlines . mconcat <$> manyTill inline (count lev $ char '=') - return $ B.header lev contents + attr <- registerHeader nullAttr contents + return $ B.headerWith attr lev contents bulletList :: MWParser Blocks bulletList = B.bulletList <$> diff --git a/tests/mediawiki-reader.native b/tests/mediawiki-reader.native index 0ab51a3aa..238413445 100644 --- a/tests/mediawiki-reader.native +++ b/tests/mediawiki-reader.native @@ -1,39 +1,39 @@ Pandoc (Meta {unMeta = fromList []}) -[Header 1 ("",[],[]) [Str "header"] -,Header 2 ("",[],[]) [Str "header",Space,Str "level",Space,Str "two"] -,Header 3 ("",[],[]) [Str "header",Space,Str "level",Space,Str "3"] -,Header 4 ("",[],[]) [Str "header",Space,Emph [Str "level"],Space,Str "four"] -,Header 5 ("",[],[]) [Str "header",Space,Str "level",Space,Str "5"] -,Header 6 ("",[],[]) [Str "header",Space,Str "level",Space,Str "6"] +[Header 1 ("header",[],[]) [Str "header"] +,Header 2 ("header-level-two",[],[]) [Str "header",Space,Str "level",Space,Str "two"] +,Header 3 ("header-level-3",[],[]) [Str "header",Space,Str "level",Space,Str "3"] +,Header 4 ("header-level-four",[],[]) [Str "header",Space,Emph [Str "level"],Space,Str "four"] +,Header 5 ("header-level-5",[],[]) [Str "header",Space,Str "level",Space,Str "5"] +,Header 6 ("header-level-6",[],[]) [Str "header",Space,Str "level",Space,Str "6"] ,Para [Str "=======",Space,Str "not",Space,Str "a",Space,Str "header",Space,Str "========"] ,Para [Code ("",[],[]) "==\160not\160a\160header\160=="] -,Header 2 ("",[],[]) [Str "emph",Space,Str "and",Space,Str "strong"] +,Header 2 ("emph-and-strong",[],[]) [Str "emph",Space,Str "and",Space,Str "strong"] ,Para [Emph [Str "emph"],Space,Strong [Str "strong"]] ,Para [Strong [Emph [Str "strong",Space,Str "and",Space,Str "emph"]]] ,Para [Strong [Emph [Str "emph",Space,Str "inside"],Space,Str "strong"]] ,Para [Strong [Str "strong",Space,Str "with",Space,Emph [Str "emph"]]] ,Para [Emph [Strong [Str "strong",Space,Str "inside"],Space,Str "emph"]] -,Header 2 ("",[],[]) [Str "horizontal",Space,Str "rule"] +,Header 2 ("horizontal-rule",[],[]) [Str "horizontal",Space,Str "rule"] ,Para [Str "top"] ,HorizontalRule ,Para [Str "bottom"] ,HorizontalRule -,Header 2 ("",[],[]) [Str "nowiki"] +,Header 2 ("nowiki",[],[]) [Str "nowiki"] ,Para [Str "''not",Space,Str "emph''"] -,Header 2 ("",[],[]) [Str "strikeout"] +,Header 2 ("strikeout",[],[]) [Str "strikeout"] ,Para [Strikeout [Str "This",Space,Str "is",Space,Emph [Str "struck",Space,Str "out"]]] -,Header 2 ("",[],[]) [Str "entities"] +,Header 2 ("entities",[],[]) [Str "entities"] ,Para [Str "hi",Space,Str "&",Space,Str "low"] ,Para [Str "hi",Space,Str "&",Space,Str "low"] ,Para [Str "G\246del"] ,Para [Str "\777\2730"] -,Header 2 ("",[],[]) [Str "comments"] +,Header 2 ("comments",[],[]) [Str "comments"] ,Para [Str "inline",Space,Str "comment"] ,Para [Str "between",Space,Str "blocks"] -,Header 2 ("",[],[]) [Str "linebreaks"] +,Header 2 ("linebreaks",[],[]) [Str "linebreaks"] ,Para [Str "hi",LineBreak,Str "there"] ,Para [Str "hi",LineBreak,Str "there"] -,Header 2 ("",[],[]) [Str ":",Space,Str "indents"] +,Header 2 ("indents",[],[]) [Str ":",Space,Str "indents"] ,Para [Str "hi"] ,DefinitionList [([], @@ -46,36 +46,36 @@ Pandoc (Meta {unMeta = fromList []}) [([], [[Plain [Str "there"]]])]]])] ,Para [Str "bud"] -,Header 2 ("",[],[]) [Str "p",Space,Str "tags"] +,Header 2 ("p-tags",[],[]) [Str "p",Space,Str "tags"] ,Para [Str "hi",Space,Str "there"] ,Para [Str "bud"] ,Para [Str "another"] -,Header 2 ("",[],[]) [Str "raw",Space,Str "html"] +,Header 2 ("raw-html",[],[]) [Str "raw",Space,Str "html"] ,Para [Str "hi",Space,RawInline (Format "html") "",Emph [Str "there"],RawInline (Format "html") "",Str "."] ,Para [RawInline (Format "html") "",Str "inserted",RawInline (Format "html") ""] ,RawBlock (Format "html") "
" ,Para [Str "hi",Space,Emph [Str "there"]] ,RawBlock (Format "html") "
" -,Header 2 ("",[],[]) [Str "sup,",Space,Str "sub,",Space,Str "del"] +,Header 2 ("sup-sub-del",[],[]) [Str "sup,",Space,Str "sub,",Space,Str "del"] ,Para [Str "H",Subscript [Str "2"],Str "O",Space,Str "base",Superscript [Emph [Str "exponent"]],Space,Strikeout [Str "hello"]] -,Header 2 ("",[],[]) [Str "inline",Space,Str "code"] +,Header 2 ("inline-code",[],[]) [Str "inline",Space,Str "code"] ,Para [Code ("",[],[]) "*\8594*",Space,Code ("",[],[]) "typed",Space,Code ("",["haskell"],[]) ">>="] -,Header 2 ("",[],[]) [Str "code",Space,Str "blocks"] +,Header 2 ("code-blocks",[],[]) [Str "code",Space,Str "blocks"] ,CodeBlock ("",[],[]) "case xs of\n (_:_) -> reverse xs\n [] -> ['*']" ,CodeBlock ("",["haskell"],[]) "case xs of\n (_:_) -> reverse xs\n [] -> ['*']" ,CodeBlock ("",["ruby","numberLines"],[("startFrom","100")]) "widgets.each do |w|\n print w.price\nend" -,Header 2 ("",[],[]) [Str "block",Space,Str "quotes"] +,Header 2 ("block-quotes",[],[]) [Str "block",Space,Str "quotes"] ,Para [Str "Regular",Space,Str "paragraph"] ,BlockQuote [Para [Str "This",Space,Str "is",Space,Str "a",Space,Str "block",Space,Str "quote."] ,Para [Str "With",Space,Str "two",Space,Str "paragraphs."]] ,Para [Str "Nother",Space,Str "paragraph."] -,Header 2 ("",[],[]) [Str "external",Space,Str "links"] +,Header 2 ("external-links",[],[]) [Str "external",Space,Str "links"] ,Para [Link [Emph [Str "Google"],Space,Str "search",Space,Str "engine"] ("http://google.com","")] ,Para [Link [Str "http://johnmacfarlane.net/pandoc/"] ("http://johnmacfarlane.net/pandoc/","")] ,Para [Link [Str "1"] ("http://google.com",""),Space,Link [Str "2"] ("http://yahoo.com","")] ,Para [Link [Str "email",Space,Str "me"] ("mailto:info@example.org","")] -,Header 2 ("",[],[]) [Str "internal",Space,Str "links"] +,Header 2 ("internal-links",[],[]) [Str "internal",Space,Str "links"] ,Para [Link [Str "Help"] ("Help","wikilink")] ,Para [Link [Str "the",Space,Str "help",Space,Str "page"] ("Help","wikilink")] ,Para [Link [Str "Helpers"] ("Help","wikilink")] @@ -83,12 +83,12 @@ Pandoc (Meta {unMeta = fromList []}) ,Para [Link [Str "Contents"] ("Help:Contents","wikilink")] ,Para [Link [Str "#My",Space,Str "anchor"] ("#My_anchor","wikilink")] ,Para [Link [Str "and",Space,Str "text"] ("Page#with_anchor","wikilink")] -,Header 2 ("",[],[]) [Str "images"] +,Header 2 ("images",[],[]) [Str "images"] ,Para [Image [Str "caption"] ("example.jpg","fig:caption")] ,Para [Image [Str "the",Space,Emph [Str "caption"],Space,Str "with",Space,Link [Str "external",Space,Str "link"] ("http://google.com","")] ("example.jpg","fig:the caption with external link")] ,Para [Image [Str "caption"] ("example.jpg","fig:caption")] ,Para [Image [Str "example.jpg"] ("example.jpg","fig:example.jpg")] -,Header 2 ("",[],[]) [Str "lists"] +,Header 2 ("lists",[],[]) [Str "lists"] ,BulletList [[Plain [Str "Start",Space,Str "each",Space,Str "line"]] ,[Plain [Str "with",Space,Str "an",Space,Str "asterisk",Space,Str "(*)."] @@ -161,10 +161,10 @@ Pandoc (Meta {unMeta = fromList []}) [[Plain [Str "Amsterdam"]] ,[Plain [Str "Rotterdam"]] ,[Plain [Str "The",Space,Str "Hague"]]] -,Header 2 ("",[],[]) [Str "math"] +,Header 2 ("math",[],[]) [Str "math"] ,Para [Str "Here",Space,Str "is",Space,Str "some",Space,Math InlineMath "x=\\frac{y^\\pi}{z}",Str "."] ,Para [Str "With",Space,Str "spaces:",Space,Math InlineMath "x=\\frac{y^\\pi}{z}",Str "."] -,Header 2 ("",[],[]) [Str "preformatted",Space,Str "blocks"] +,Header 2 ("preformatted-blocks",[],[]) [Str "preformatted",Space,Str "blocks"] ,Para [Code ("",[],[]) "Start\160each\160line\160with\160a\160space.",LineBreak,Code ("",[],[]) "Text\160is\160",Strong [Code ("",[],[]) "preformatted"],Code ("",[],[]) "\160and",LineBreak,Emph [Code ("",[],[]) "markups"],Code ("",[],[]) "\160",Strong [Emph [Code ("",[],[]) "can"]],Code ("",[],[]) "\160be\160done."] ,Para [Code ("",[],[]) "\160hell\160\160\160\160\160\160yeah"] ,Para [Code ("",[],[]) "Start\160with\160a\160space\160in\160the\160first\160column,",LineBreak,Code ("",[],[]) "(before\160the\160).",LineBreak,Code ("",[],[]) "",LineBreak,Code ("",[],[]) "Then\160your\160block\160format\160will\160be",LineBreak,Code ("",[],[]) "\160\160\160\160maintained.",LineBreak,Code ("",[],[]) "",LineBreak,Code ("",[],[]) "This\160is\160good\160for\160copying\160in\160code\160blocks:",LineBreak,Code ("",[],[]) "",LineBreak,Code ("",[],[]) "def\160function():",LineBreak,Code ("",[],[]) "\160\160\160\160\"\"\"documentation\160string\"\"\"",LineBreak,Code ("",[],[]) "",LineBreak,Code ("",[],[]) "\160\160\160\160if\160True:",LineBreak,Code ("",[],[]) "\160\160\160\160\160\160\160\160print\160True",LineBreak,Code ("",[],[]) "\160\160\160\160else:",LineBreak,Code ("",[],[]) "\160\160\160\160\160\160\160\160print\160False"] @@ -174,12 +174,12 @@ Pandoc (Meta {unMeta = fromList []}) ,Para [Str "Don't",Space,Str "need"] ,Para [Code ("",[],[]) "a\160blank\160line"] ,Para [Str "around",Space,Str "a",Space,Str "preformatted",Space,Str "block."] -,Header 2 ("",[],[]) [Str "templates"] +,Header 2 ("templates",[],[]) [Str "templates"] ,RawBlock (Format "mediawiki") "{{Welcome}}" ,RawBlock (Format "mediawiki") "{{Foo:Bar}}" ,RawBlock (Format "mediawiki") "{{Thankyou|all your effort|Me}}" ,Para [Str "Written",Space,RawInline (Format "mediawiki") "{{{date}}}",Space,Str "by",Space,RawInline (Format "mediawiki") "{{{name}}}",Str "."] -,Header 2 ("",[],[]) [Str "tables"] +,Header 2 ("tables",[],[]) [Str "tables"] ,Table [] [AlignDefault,AlignDefault] [0.0,0.0] [[] ,[]] @@ -245,6 +245,6 @@ Pandoc (Meta {unMeta = fromList []}) [[]] [[[Para [Str "Orange"]]]] ,Para [Str "Paragraph",Space,Str "after",Space,Str "the",Space,Str "table."] -,Header 2 ("",[],[]) [Str "notes"] +,Header 2 ("notes",[],[]) [Str "notes"] ,Para [Str "My",Space,Str "note!",Note [Plain [Str "This."]]] ,Para [Str "URL",Space,Str "note.",Note [Plain [Link [Str "http://docs.python.org/library/functions.html#range"] ("http://docs.python.org/library/functions.html#range","")]]]] -- cgit v1.2.3 From e690c87dc4c3ff79689d9a887a9b5e3bffbb5d37 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 17 Nov 2013 09:07:25 -0800 Subject: LaTeX reader: Support `\textnormal` as span with class "nodecor". This is needed for pandoc-citeproc. --- src/Text/Pandoc/Readers/LaTeX.hs | 1 + 1 file changed, 1 insertion(+) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 681dcb077..689b12c8e 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -373,6 +373,7 @@ inlineCommands = M.fromList $ , ("backslash", lit "\\") , ("slash", lit "/") , ("textbf", strong <$> tok) + , ("textnormal", spanWith ("",["nodecor"],[]) <$> tok) , ("ldots", lit "…") , ("dots", lit "…") , ("mdots", lit "…") -- cgit v1.2.3 From d07dc971da0123149e751dff987cf9df1a55b75d Mon Sep 17 00:00:00 2001 From: "Shaun Attfield (shaun@victor)" Date: Mon, 18 Nov 2013 11:53:13 +0200 Subject: Epub Writer: Add cover reference to guide element Fixes an issue with calibre http://calibre-ebook.com/ putting the cover at the end of the book if the spine has linear="no". Apparently this is best practice for other converters as well. http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm#Section2.6 --- src/Text/Pandoc/Writers/EPUB.hs | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index ac0e7610c..ca69a0fd4 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -260,8 +260,10 @@ 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")] $ () ] ] let contentsEntry = mkEntry "content.opf" contentsData -- cgit v1.2.3 From a3eba6ee848497f98f72aef5dad112c49bdd4fec Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Mon, 18 Nov 2013 20:28:27 -0800 Subject: LaTeX reader: Parse contents of curly quotes or matched `"` as quotes. --- src/Text/Pandoc/Readers/LaTeX.hs | 24 +++++++++++++++++------- tests/latex-reader.native | 2 +- 2 files changed, 18 insertions(+), 8 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 689b12c8e..75e29ebb9 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -163,13 +163,23 @@ mathChars = concat <$> <|> (\c -> ['\\',c]) <$> (try $ char '\\' *> anyChar) ) +quoted' :: (Inlines -> Inlines) -> LP String -> LP () -> LP Inlines +quoted' f starter ender = do + startchs <- starter + try ((f . mconcat) <$> manyTill inline ender) <|> lit startchs + double_quote :: LP Inlines -double_quote = (doubleQuoted . mconcat) <$> - (try $ string "``" *> manyTill inline (try $ string "''")) +double_quote = + ( quoted' doubleQuoted (try $ string "``") (void $ try $ string "''") + <|> quoted' doubleQuoted (string "“") (void $ char '”') + <|> quoted' doubleQuoted (string "\"") (void $ char '"') + ) single_quote :: LP Inlines -single_quote = (singleQuoted . mconcat) <$> - (try $ char '`' *> manyTill inline (try $ char '\'' >> notFollowedBy letter)) +single_quote = + ( quoted' singleQuoted (string "`") (try $ char '\'' >> notFollowedBy letter) + <|> quoted' singleQuoted (string "‘") (try $ char '’' >> notFollowedBy letter) + ) inline :: LP Inlines inline = (mempty <$ comment) @@ -181,10 +191,10 @@ inline = (mempty <$ comment) ((char '-') *> option (str "–") (str "—" <$ char '-'))) <|> double_quote <|> single_quote - <|> (str "“" <$ try (string "``")) -- nb. {``} won't be caught by double_quote <|> (str "”" <$ try (string "''")) - <|> (str "‘" <$ char '`') -- nb. {`} won't be caught by single_quote + <|> (str "”" <$ char '”') <|> (str "’" <$ char '\'') + <|> (str "’" <$ char '’') <|> (str "\160" <$ char '~') <|> (mathDisplay $ string "$$" *> mathChars <* string "$$") <|> (mathInline $ char '$' *> mathChars <* char '$') @@ -755,7 +765,7 @@ inlineText :: LP Inlines inlineText = str <$> many1 inlineChar inlineChar :: LP Char -inlineChar = noneOf "\\$%^_&~#{}^'`-[] \t\n" +inlineChar = noneOf "\\$%^_&~#{}^'`\"‘’“”-[] \t\n" environment :: LP Blocks environment = do diff --git a/tests/latex-reader.native b/tests/latex-reader.native index 15b667b2f..fcc3153cf 100644 --- a/tests/latex-reader.native +++ b/tests/latex-reader.native @@ -302,7 +302,7 @@ Pandoc (Meta {unMeta = fromList [("authors",MetaList [MetaInlines [Str "John",Sp ,Para [Str "4",Space,Str "<",Space,Str "5."] ,Para [Str "6",Space,Str ">",Space,Str "5."] ,Para [Str "Backslash:",Space,Str "\\"] -,Para [Str "Backtick:",Space,Str "\8216"] +,Para [Str "Backtick:",Space,Str "`"] ,Para [Str "Asterisk:",Space,Str "*"] ,Para [Str "Underscore:",Space,Str "_"] ,Para [Str "Left",Space,Str "brace:",Space,Str "{"] -- cgit v1.2.3 From 5b27480e54f4ecfc145bde133f1f51865d9b12a0 Mon Sep 17 00:00:00 2001 From: "Shaun Attfield (shaun@victor)" Date: Tue, 19 Nov 2013 08:20:27 +0200 Subject: Epub Writer: Add cover reference to guide element (v2) Avoiding an unnecessary list concatenation. Fixes an issue with calibre http://calibre-ebook.com/ putting the cover at the end of the book if the spine has linear="no". Apparently this is best practice for other converters as well. http://www.idpf.org/epub/20/spec/OPF_2.0.1_draft.htm#Section2.6 --- src/Text/Pandoc/Writers/EPUB.hs | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index ca69a0fd4..be8de7073 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -260,10 +260,11 @@ 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","cover"),("title","Cover"),("href","cover.xhtml")] $ () ] + [ unode "reference" ! + [("type","toc"),("title",plainTitle),("href","nav.xhtml")] $ () + , unode "reference" ! + [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () + ] ] let contentsEntry = mkEntry "content.opf" contentsData -- cgit v1.2.3 From c226a57eaa7f1e9d4782c7ea3b56e72251a6487a Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 19 Nov 2013 12:01:52 -0800 Subject: Docbook writer: Hierarchicalize block content in metadata. Previously headers just disappeared from block-level metadata when it was used in templates. Now we apply the 'hierarchicalize' transformation. Note that a block headed by a level-2 header will turn into a `` element. --- src/Text/Pandoc/Writers/Docbook.hs | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'src/Text') 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 -- cgit v1.2.3 From e290d91c93b265ce7763797a5c995d29fb683c67 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 19 Nov 2013 13:09:34 -0800 Subject: MIME: In looking up extensions, drop the encoding info. E.g. for 'image/jpg;base64' we should lookup 'image/jpg'. --- src/Text/Pandoc/MIME.hs | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/MIME.hs b/src/Text/Pandoc/MIME.hs index d9cb94a33..1f5f6f862 100644 --- a/src/Text/Pandoc/MIME.hs +++ b/src/Text/Pandoc/MIME.hs @@ -40,7 +40,8 @@ getMimeType f = M.lookup (map toLower $ drop 1 $ takeExtension f) mimeTypes where mimeTypes = M.fromList mimeTypesList extensionFromMimeType :: String -> Maybe String -extensionFromMimeType mimetype = M.lookup mimetype reverseMimeTypes +extensionFromMimeType mimetype = M.lookup (takeWhile (/=';') mimetype) reverseMimeTypes + -- note: we just look up the basic mime type, dropping the content-encoding etc. where reverseMimeTypes = M.fromList $ map (\(k,v) -> (v,k)) mimeTypesList mimeTypesList :: [(String, String)] -- cgit v1.2.3 From 83b9a66bf468c0f32bd96be92bb571bd83b73903 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 19 Nov 2013 13:15:24 -0800 Subject: Shared: Fixed bug in openURL with data: URIs. Previously the base-64 encoded bytestring was returned. We now decode it so it's a proper image! This should fix parsing of data: URLs. --- src/Text/Pandoc/Shared.hs | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index d6ccdae66..8dcd88148 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -108,6 +108,7 @@ import Text.HTML.TagSoup (renderTagsOptions, RenderOptions(..), Tag(..), import qualified Data.ByteString as BS import qualified Data.ByteString.Char8 as B8 import Text.Pandoc.Compat.Monoid +import Data.ByteString.Base64 (decodeLenient) #ifdef EMBED_DATA_FILES import Text.Pandoc.Data (dataFiles) @@ -641,7 +642,7 @@ openURL u | "data:" `isPrefixOf` u = let mime = takeWhile (/=',') $ drop 5 u contents = B8.pack $ unEscapeString $ drop 1 $ dropWhile (/=',') u - in return $ Right (contents, Just mime) + in return $ Right (decodeLenient contents, Just mime) #ifdef HTTP_CONDUIT | otherwise = E.try $ do req <- parseUrl u -- cgit v1.2.3 From 3d453f096cfa12e231c5a4d4c8e468378e20e5e8 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 19 Nov 2013 13:16:31 -0800 Subject: Docx writer: Use mime type info returned by fetchItem. --- src/Text/Pandoc/Writers/Docx.hs | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 5d1647844..b9c198a78 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -55,8 +55,8 @@ 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] @@ -737,7 +737,7 @@ 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 @@ -776,19 +776,21 @@ 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) -- cgit v1.2.3 From cf149fcf38d98b1bee79ecd9056fa0f46264e7ce Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 22 Nov 2013 19:41:08 -0800 Subject: Fixed bug with intraword emphasis. Closes #1066. --- src/Text/Pandoc/Readers/Markdown.hs | 3 ++- tests/Tests/Readers/Markdown.hs | 5 +++++ 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index ea49d8c1d..33d1a9620 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1455,6 +1455,7 @@ enclosure c = do -- Parse inlines til you hit one c or a sequence of two cs. -- If one c, emit emph and then parse two. -- If two cs, emit strong and then parse one. +-- Otherwise, emit ccc then the results. three :: Char -> MarkdownParser (F Inlines) three c = do contents <- mconcat <$> many (notFollowedBy (char c) >> inline) @@ -1479,7 +1480,7 @@ one c prefix' = do contents <- mconcat <$> many ( (notFollowedBy (char c) >> inline) <|> try (string [c,c] >> notFollowedBy (char c) >> - two c prefix') ) + two c mempty) ) (char c >> return (B.emph <$> (prefix' <> contents))) <|> return (return (B.str [c]) <> (prefix' <> contents)) diff --git a/tests/Tests/Readers/Markdown.hs b/tests/Tests/Readers/Markdown.hs index ccca147ab..b04ff9a0d 100644 --- a/tests/Tests/Readers/Markdown.hs +++ b/tests/Tests/Readers/Markdown.hs @@ -136,6 +136,11 @@ tests = [ testGroup "inline code" "`*` {.haskell .special x=\"7\"}" =?> para (codeWith ("",["haskell","special"],[("x","7")]) "*") ] + , testGroup "emph and strong" + [ "two strongs in emph" =: + "***a**b **c**d*" =?> para (emph (strong (str "a") <> str "b" <> space + <> strong (str "c") <> str "d")) + ] , testGroup "raw LaTeX" [ "in URL" =: "\\begin\n" =?> para (text "\\begin") -- cgit v1.2.3 From 526762bf222dbab199f6ff90c925fe18535c698f Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 22 Nov 2013 19:51:07 -0800 Subject: ConTeXt writer: Use setupcaption to separate style from content. Instead of adding 'nunumber' every time we place a figure... Closes #1067. --- data/templates | 2 +- src/Text/Pandoc/Writers/ConTeXt.hs | 2 +- tests/writer.context | 4 +++- 3 files changed, 5 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/data/templates b/data/templates index 4bdebc73b..8cadd4f20 160000 --- a/data/templates +++ b/data/templates @@ -1 +1 @@ -Subproject commit 4bdebc73b0b2025cf01704e7e564088c34d8f86c +Subproject commit 8cadd4f2044c0c25842eeb5a2370a6e3384f4bd4 diff --git a/src/Text/Pandoc/Writers/ConTeXt.hs b/src/Text/Pandoc/Writers/ConTeXt.hs index 0379f8b0a..179d9bc5b 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[here]" <> braces capt <> braces ("\\externalfigure" <> brackets (text src)) <> blankline blockToConTeXt (Para lst) = do contents <- inlineListToConTeXt lst diff --git a/tests/writer.context b/tests/writer.context index 114d00b3c..fb95f5615 100644 --- a/tests/writer.context +++ b/tests/writer.context @@ -30,6 +30,8 @@ \setupitemize[autointro] % prevent orphan list intro \setupitemize[indentnext=no] +\setupcaption[figure][number=no] % don't number figures + \setupthinrules[width=15em] % width of horizontal rules \setupdelimitedtext @@ -842,7 +844,7 @@ or here: From \quotation{Voyage dans la Lune} by Georges Melies (1902): -\placefigure[here,nonumber]{lalune}{\externalfigure[lalune.jpg]} +\placefigure[here]{lalune}{\externalfigure[lalune.jpg]} Here is a movie {\externalfigure[movie.jpg]} icon. -- cgit v1.2.3 From 4321a09b7fa5b52e72aef4fa637392d7da4eb808 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 22 Nov 2013 23:36:19 -0800 Subject: EPUB writer: Ensure that same identifier is used throughout. If dc:identifier is given in metadata, we use that; otherwise we use a random uuid. Closes #1044. --- src/Text/Pandoc/Writers/EPUB.hs | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index be8de7073..98f043cd8 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -198,7 +198,12 @@ writeEPUB opts doc@(Pandoc meta _) = do let lang = case lookup "lang" (writerVariables opts') of Just x -> x Nothing -> localeLang - uuid <- getRandomUUID + let userNodes = onlyElems $ parseXML $ writerEpubMetadata opts' + let mbIdent = findElement (QName "identifier" Nothing (Just "dc")) + $ unode "dummy" ! [] $ userNodes + uuid <- case mbIdent of + Just id' -> return $ trim $ strContent id' + Nothing -> fmap show getRandomUUID let chapterNode ent = unode "item" ! ([("id", takeBaseName $ eRelativePath ent), ("href", eRelativePath ent), @@ -230,7 +235,7 @@ writeEPUB opts doc@(Pandoc meta _) = do EPUB3 -> "3.0") ,("xmlns","http://www.idpf.org/2007/opf") ,("unique-identifier","BookId")] $ - [ metadataElement version (writerEpubMetadata opts') + [ metadataElement version userNodes uuid lang plainTitle plainAuthors plainDate currentTime mbCoverImage , unode "manifest" $ [ unode "item" ! [("id","ncx"), ("href","toc.ncx") @@ -311,7 +316,7 @@ writeEPUB opts doc@(Pandoc meta _) = do ,("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") @@ -384,13 +389,12 @@ writeEPUB opts doc@(Pandoc meta _) = do (picEntries ++ cpicEntry ++ cpgEntry ++ chapterEntries ++ fontEntries)) return $ fromArchive archive -metadataElement :: EPUBVersion -> String -> UUID -> String -> String -> [String] +metadataElement :: EPUBVersion -> [Element] -> String -> 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/") +metadataElement version userNodes uuid lang title authors date currentTime mbCoverImage = + let elt = unode "metadata" ! [("xmlns:dc","http://purl.org/dc/elements/1.1/") ,("xmlns:opf","http://www.idpf.org/2007/opf")] $ - filter isMetadataElement $ onlyElems userNodes + filter isMetadataElement userNodes dublinElements = ["contributor","coverage","creator","date", "description","format","identifier","language","publisher", "relation","rights","source","subject","title","type"] @@ -401,7 +405,7 @@ metadataElement version metadataXML uuid lang title authors date currentTime mbC 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 | + [ unode "dc:identifier" ! [("id","BookId")] $ uuid | not (elt `contains` "identifier") ] ++ [ unode "dc:creator" ! [("opf:role","aut") | version == EPUB2] $ a | a <- authors, not (elt `contains` "creator") ] ++ -- cgit v1.2.3 From 56277bacea73818a7541b60987e5bc877782baab Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 22 Nov 2013 23:40:46 -0800 Subject: EPUB writer: Don't include node for cover.xhtml if no cover! --- src/Text/Pandoc/Writers/EPUB.hs | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 98f043cd8..ef4c7be23 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -267,8 +267,9 @@ writeEPUB opts doc@(Pandoc meta _) = do , unode "guide" $ [ unode "reference" ! [("type","toc"),("title",plainTitle),("href","nav.xhtml")] $ () - , unode "reference" ! - [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () + ] ++ + [ unode "reference" ! + [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () | mbCoverImage /= Nothing ] ] let contentsEntry = mkEntry "content.opf" contentsData -- cgit v1.2.3 From c1ff65e5ef0b5b8d78149cf666cda1de4a0e4fdb Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 23 Nov 2013 14:24:33 -0800 Subject: HTML writer: Handle csl flipflopping spans (csl-no-emph, etc.) --- src/Text/Pandoc/Writers/HTML.hs | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index 424843539..641652276 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -613,8 +613,22 @@ 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 -- cgit v1.2.3 From b82ef0e29a652b7e1629b6e2e5f56140ed08e633 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 23 Nov 2013 14:41:22 -0800 Subject: LaTeX writer: Handle csl flipflopping spans (csl-no-emph, etc.) --- src/Text/Pandoc/Writers/LaTeX.hs | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index 7ff64cf74..f3cbcf19f 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -602,7 +602,16 @@ 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) = -- cgit v1.2.3 From e1a9a61774cce807c6db2fff7b8609b2d9dbc678 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 23 Nov 2013 14:52:14 -0800 Subject: Docx writer: Implemented csl flipflopping spans. --- src/Text/Pandoc/Writers/Docx.hs | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index b9c198a78..5c7341b69 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -638,7 +638,12 @@ 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) = -- cgit v1.2.3 From 659596876b74a2ce99ec5f7253962b27ca091354 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 29 Nov 2013 18:01:07 -0800 Subject: EPUB writer: Improved metadata handling. * Metadata may now be included in YAML blocks in a markdown document. For example, --- title: - type: main text: My Book - type: subtitle text: An investigation of metadata creator: - role: author text: John Smith - role: editor text: Sarah Jones identifier: - scheme: DOI text: doi:10.234234.234/33 publisher: My Press rights: (c) 2007 John Smith, CC BY-NC ... * Metadata may still be provided using `--epub-metadata`; it will be merged with the metadata in YAML blocks. * meta tags are used instead of opf attributes for EPUB3. --- src/Text/Pandoc/Writers/EPUB.hs | 642 ++++++++++++++++++++++++++++++++++++---- 1 file changed, 591 insertions(+), 51 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index ef4c7be23..4f1de5df0 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 @@ -30,7 +30,8 @@ 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) @@ -40,6 +41,7 @@ import qualified Data.ByteString.Lazy.Char8 as B8 import Text.Pandoc.UTF8 ( fromStringLazy, toString ) import Text.Pandoc.SelfContained ( makeSelfContained ) import Codec.Archive.Zip +import Control.Applicative ((<$>)) import Data.Time.Clock.POSIX import Data.Time import System.Locale @@ -70,6 +72,216 @@ 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 + } 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 t] + +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 = maybe "" id $ 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 + } + where identifiers = getIdentifier meta + titles = getTitle meta + date = maybe "" id $ + (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 = fmap (const "cover-image") $ + lookup "epub-cover-image" $ writerVariables opts + -- | Produce an EPUB file from a Pandoc document. writeEPUB :: WriterOptions -- ^ Writer options -> Pandoc -- ^ Document to convert @@ -192,18 +404,6 @@ writeEPUB opts doc@(Pandoc meta _) = do let containsMathML ent = " 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 - let userNodes = onlyElems $ parseXML $ writerEpubMetadata opts' - let mbIdent = findElement (QName "identifier" Nothing (Just "dc")) - $ unode "dummy" ! [] $ userNodes - uuid <- case mbIdent of - Just id' -> return $ trim $ strContent id' - Nothing -> fmap show getRandomUUID let chapterNode ent = unode "item" ! ([("id", takeBaseName $ eRelativePath ent), ("href", eRelativePath ent), @@ -221,22 +421,23 @@ writeEPUB opts doc@(Pandoc meta _) = do [("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 + metadata <- getEPUBMetadata opts' meta + 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 $ unode "package" ! [("version", case version of EPUB2 -> "2.0" EPUB3 -> "3.0") ,("xmlns","http://www.idpf.org/2007/opf") - ,("unique-identifier","BookId")] $ - [ metadataElement version userNodes - 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")] $ () @@ -266,7 +467,8 @@ writeEPUB opts doc@(Pandoc meta _) = do map chapterRefNode chapterEntries) , unode "guide" $ [ unode "reference" ! - [("type","toc"),("title",plainTitle),("href","nav.xhtml")] $ () + [("type","toc"),("title",plainTitle), + ("href","nav.xhtml")] $ () ] ++ [ unode "reference" ! [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () | mbCoverImage /= Nothing @@ -390,32 +592,96 @@ writeEPUB opts doc@(Pandoc meta _) = do (picEntries ++ cpicEntry ++ cpgEntry ++ chapterEntries ++ fontEntries)) return $ fromArchive archive -metadataElement :: EPUBVersion -> [Element] -> String -> String -> String -> [String] - -> String -> UTCTime -> Maybe a -> Element -metadataElement version userNodes uuid lang title authors date currentTime mbCoverImage = - let elt = unode "metadata" ! [("xmlns:dc","http://purl.org/dc/elements/1.1/") - ,("xmlns:opf","http://www.idpf.org/2007/opf")] $ - filter isMetadataElement 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")] $ 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 [] + (\ci -> [unode "meta" ! [("name","cover"), ("content",ci)] $ ()]) + $ 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" @@ -527,3 +793,277 @@ replaceRefs refTable = walk replaceOneRef Just url -> Link lab (url,tit) Nothing -> x replaceOneRef x = x + +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") + ] + -- cgit v1.2.3 From 96b678d823b544e3da7dd8531615c52de6164cb8 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 15:17:38 -0800 Subject: Allow specification of epub-cover-image in YAML metadata. --- README | 6 +++++- src/Text/Pandoc/Writers/EPUB.hs | 17 ++++++++--------- 2 files changed, 13 insertions(+), 10 deletions(-) (limited to 'src/Text') diff --git a/README b/README index 3afbe8b0b..35194769b 100644 --- a/README +++ b/README @@ -541,7 +541,9 @@ Options affecting specific writers `--epub-cover-image=`*FILE* : Use the specified image as the EPUB cover. It is recommended - that the image be less than 1000px in width and height. + that the image be less than 1000px in width and height. Note that + in a markdown source document you can also specify `epub-cover-image` + in a YAML metadata block (see [EPUB Metadata], below). `--epub-metadata=`*FILE* : Look in the specified XML file for metadata for the EPUB. @@ -2818,6 +2820,8 @@ The following fields are recognized: ~ A string value. `rights` ~ A string value. +`epub-cover-image` + ~ A string value (path to cover image). Literate Haskell support ======================== diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 4f1de5df0..a3b01848e 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -279,8 +279,7 @@ metadataFromMeta opts meta = EPUBMetadata{ relation = metaValueToString <$> lookupMeta "relation" meta coverage = metaValueToString <$> lookupMeta "coverage" meta rights = metaValueToString <$> lookupMeta "rights" meta - coverImage = fmap (const "cover-image") $ - lookup "epub-cover-image" $ writerVariables opts + coverImage = lookup "epub-cover-image" (writerVariables opts) -- | Produce an EPUB file from a Pandoc document. writeEPUB :: WriterOptions -- ^ Writer options @@ -305,11 +304,11 @@ 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 @@ -421,7 +420,6 @@ writeEPUB opts doc@(Pandoc meta _) = do [("id", takeBaseName $ eRelativePath ent), ("href", eRelativePath ent), ("media-type", maybe "" id $ getMimeType $ eRelativePath ent)] $ () - metadata <- getEPUBMetadata opts' meta let plainTitle = case docTitle meta of [] -> case epubTitle metadata of [] -> "UNTITLED" @@ -452,7 +450,7 @@ writeEPUB opts doc@(Pandoc meta _) = do map pictureNode (cpicEntry ++ picEntries) ++ map fontNode fontEntries , unode "spine" ! [("toc","ncx")] $ - case mbCoverImage of + case epubCoverImage metadata of Nothing -> [] Just _ -> [ unode "itemref" ! [("idref", "cover"),("linear","no")] $ () ] @@ -471,7 +469,7 @@ writeEPUB opts doc@(Pandoc meta _) = do ("href","nav.xhtml")] $ () ] ++ [ unode "reference" ! - [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () | mbCoverImage /= Nothing + [("type","cover"),("title","Cover"),("href","cover.xhtml")] $ () | epubCoverImage metadata /= Nothing ] ] let contentsEntry = mkEntry "content.opf" contentsData @@ -526,7 +524,7 @@ writeEPUB opts doc@(Pandoc meta _) = do ,("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")] $ ()] @@ -623,7 +621,8 @@ metadataElement version md currentTime = coverageNodes = maybe [] (dcTag' "coverage") $ epubCoverage md rightsNodes = maybe [] (dcTag' "rights") $ epubRights md coverImageNodes = maybe [] - (\ci -> [unode "meta" ! [("name","cover"), ("content",ci)] $ ()]) + (const $ [unode "meta" ! [("name","cover"), + ("content","cover-image")] $ ()]) $ epubCoverImage md modifiedNodes = [ unode "meta" ! [("property", "dcterms:modified")] $ (showDateTimeISO8601 currentTime) | version == EPUB3 ] -- cgit v1.2.3 From 17ef39d1577a7c844f8747294f0e7a0486c1edaa Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 15:25:28 -0800 Subject: Fixed `cover-image` in EPUB YAML metadata. --- README | 4 ++-- src/Text/Pandoc/Writers/EPUB.hs | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/README b/README index 35194769b..2a5ecc6ad 100644 --- a/README +++ b/README @@ -542,7 +542,7 @@ Options affecting specific writers `--epub-cover-image=`*FILE* : Use the specified image as the EPUB cover. It is recommended that the image be less than 1000px in width and height. Note that - in a markdown source document you can also specify `epub-cover-image` + in a markdown source document you can also specify `cover-image` in a YAML metadata block (see [EPUB Metadata], below). `--epub-metadata=`*FILE* @@ -2820,7 +2820,7 @@ The following fields are recognized: ~ A string value. `rights` ~ A string value. -`epub-cover-image` +`cover-image` ~ A string value (path to cover image). Literate Haskell support diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index a3b01848e..9745db5a0 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -279,7 +279,8 @@ metadataFromMeta opts meta = EPUBMetadata{ relation = metaValueToString <$> lookupMeta "relation" meta coverage = metaValueToString <$> lookupMeta "coverage" meta rights = metaValueToString <$> lookupMeta "rights" meta - coverImage = lookup "epub-cover-image" (writerVariables opts) + coverImage = lookup "epub-cover-image" (writerVariables opts) `mplus` + (metaValueToString <$> lookupMeta "cover-image" meta) -- | Produce an EPUB file from a Pandoc document. writeEPUB :: WriterOptions -- ^ Writer options -- cgit v1.2.3 From bb0f299165de22e119675aa6ceed0192a80c78e8 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 16:16:35 -0800 Subject: ConTeXt writer: Don't hardcode figure/table placement. Instead, let this be set in the template, using `\setupfloat`. Thanks to Aditya Mahajan for the suggestion. --- data/templates | 2 +- src/Text/Pandoc/Writers/ConTeXt.hs | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'src/Text') diff --git a/data/templates b/data/templates index 8cadd4f20..713a8f63d 160000 --- a/data/templates +++ b/data/templates @@ -1 +1 @@ -Subproject commit 8cadd4f2044c0c25842eeb5a2370a6e3384f4bd4 +Subproject commit 713a8f63d5589ab9313869e47b03cf7f49e00e98 diff --git a/src/Text/Pandoc/Writers/ConTeXt.hs b/src/Text/Pandoc/Writers/ConTeXt.hs index 179d9bc5b..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 $$ "\\placefigure[here]" <> braces capt <> + return $ blankline $$ "\\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 $$ -- cgit v1.2.3 From 7aa4d519686af1416eaf3b380f8584ab89569c41 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 17:00:58 -0800 Subject: ODT writer: Add `draw:name` attribute to `draw:frame` elements. This is reported to be necessary to avoid an error from recent versions of Libre Office when files contain more than one image. Closes #1069. Thanks to wmanley for reporting and diagnosing the problem. --- src/Text/Pandoc/Writers/OpenDocument.hs | 10 ++++++++-- tests/writer.opendocument | 4 ++-- 2 files changed, 10 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/OpenDocument.hs b/src/Text/Pandoc/Writers/OpenDocument.hs index 565f5f869..4ddfd7166 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 @@ -380,7 +382,7 @@ inlineToOpenDocument o ils then return $ preformatted 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 +391,11 @@ 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/tests/writer.opendocument b/tests/writer.opendocument index b3888e34d..81c793a62 100644 --- a/tests/writer.opendocument +++ b/tests/writer.opendocument @@ -1576,9 +1576,9 @@ link in pointy braces. Images From “Voyage dans la Lune” by Georges Melies (1902): - + Here is a movie - + icon. Footnotes -- cgit v1.2.3 From 7f09c1834da9f87e7715f5c9dc52f4b730da8f3f Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 17:59:28 -0800 Subject: Markdown writer: Fix rendering of tight sublists. E.g. - foo - bar - baz Previously a spurious blank line was included before the last item. Closes #1050. --- src/Text/Pandoc/Writers/Markdown.hs | 9 ++++++++- tests/Tests/Writers/Markdown.hs | 4 ++++ 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Markdown.hs b/src/Text/Pandoc/Writers/Markdown.hs index eefcd547a..60d474263 100644 --- a/src/Text/Pandoc/Writers/Markdown.hs +++ b/src/Text/Pandoc/Writers/Markdown.hs @@ -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 diff --git a/tests/Tests/Writers/Markdown.hs b/tests/Tests/Writers/Markdown.hs index 99b85dfb7..c2a8f5903 100644 --- a/tests/Tests/Writers/Markdown.hs +++ b/tests/Tests/Writers/Markdown.hs @@ -31,4 +31,8 @@ tests :: [Test] tests = [ "indented code after list" =: (orderedList [ para "one" <> para "two" ] <> codeBlock "test") =?> "1. one\n\n two\n\n\n\n test" + , "list with tight sublist" + =: bulletList [ plain "foo" <> bulletList [ plain "bar" ], + plain "baz" ] + =?> "- foo\n - bar\n- baz\n" ] -- cgit v1.2.3 From 37569f30ed35a45b4765e9cd6c408cd167dd3452 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 19:33:02 -0800 Subject: EPUB writer: Add properties attribute to cover-image item for EPUB v3. --- src/Text/Pandoc/Writers/EPUB.hs | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 9745db5a0..6e1a391af 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -448,7 +448,12 @@ 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 epubCoverImage metadata of -- cgit v1.2.3 From 4501344d45947bee79707522efdc77bfdc650bd6 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 19:43:12 -0800 Subject: MIME: Add entry for jfif. --- src/Text/Pandoc/MIME.hs | 1 + 1 file changed, 1 insertion(+) (limited to 'src/Text') diff --git a/src/Text/Pandoc/MIME.hs b/src/Text/Pandoc/MIME.hs index 1f5f6f862..f41aa98bb 100644 --- a/src/Text/Pandoc/MIME.hs +++ b/src/Text/Pandoc/MIME.hs @@ -220,6 +220,7 @@ mimeTypesList = -- List borrowed from happstack-server. ,("jnlp","application/x-java-jnlp-file") ,("jpe","image/jpeg") ,("jpeg","image/jpeg") + ,("jfif","image/jpeg") ,("jpg","image/jpeg") ,("js","application/x-javascript") ,("kar","audio/midi") -- cgit v1.2.3 From a7067ab22b0ce297ccc5271fca59b9f9e787195e Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 19:43:24 -0800 Subject: EPUB writer: Simplify imageTypeOf using getMimeType. --- src/Text/Pandoc/Writers/EPUB.hs | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 6e1a391af..7be7eb368 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -736,15 +736,9 @@ ppTopElement = ("\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, -- cgit v1.2.3 From 0e158d8d8dc277f19d1f7d2686c2672f08410f82 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 30 Nov 2013 20:20:18 -0800 Subject: EPUB writer: Insert "svg" property as needed in opf (EPUB 3). --- src/Text/Pandoc/Writers/EPUB.hs | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 7be7eb368..b23f43a6e 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -401,15 +401,21 @@ writeEPUB opts doc@(Pandoc meta _) = do let chapterEntries = zipWith chapToEntry [1..] chapters -- incredibly inefficient (TODO): - let containsMathML ent = " [] + xs -> [("properties", unwords xs)]) + $ () let chapterRefNode ent = unode "itemref" ! [("idref", takeBaseName $ eRelativePath ent)] $ () let pictureNode ent = unode "item" ! -- cgit v1.2.3 From 82813b55852c99c2e4d179083c119937c39d5398 Mon Sep 17 00:00:00 2001 From: Shaun Attfield Date: Sun, 1 Dec 2013 10:19:08 +0200 Subject: normalizeDate: Allow dates with year only (%Y) --- src/Text/Pandoc/Shared.hs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index 8dcd88148..7592b7659 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -270,7 +270,7 @@ normalizeDate s = fmap (formatTime defaultTimeLocale "%F") (msum $ map (\fs -> parsetimeWith fs s) formats :: Maybe Day) where parsetimeWith = parseTime defaultTimeLocale formats = ["%x","%m/%d/%Y", "%D","%F", "%d %b %Y", - "%d %B %Y", "%b. %d, %Y", "%B %d, %Y"] + "%d %B %Y", "%b. %d, %Y", "%B %d, %Y", "%Y"] -- -- Pandoc block and inline list processing -- cgit v1.2.3 From 1f6238f3ba1999988fbb2b25b091171451647d4e Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 1 Dec 2013 10:16:56 -0800 Subject: EPUB writer: Allow partial dates: YYYY, YYYY-MM. Improves on #1074, since now we don't default to January 1. --- src/Text/Pandoc/Writers/EPUB.hs | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index b23f43a6e..59c90b16d 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -56,7 +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) @@ -171,7 +171,8 @@ addMetadataFromXML e@(Element (QName name _ (Just "dc")) attrs _ _) md , titleFileAs = getAttr "file-as" , titleType = getAttr "type" } : epubTitle md } - | name == "date" = md{ epubDate = maybe "" id $ normalizeDate $ strContent e } + | name == "date" = md{ epubDate = maybe "" id $ normalizeDate' + $ strContent e } | name == "language" = md{ epubLanguage = strContent e } | name == "creator" = md{ epubCreator = Creator{ creatorText = strContent e @@ -265,7 +266,7 @@ metadataFromMeta opts meta = EPUBMetadata{ where identifiers = getIdentifier meta titles = getTitle meta date = maybe "" id $ - (metaValueToString <$> lookupMeta "date" meta) >>= normalizeDate + (metaValueToString <$> lookupMeta "date" meta) >>= normalizeDate' language = maybe "" metaValueToString $ lookupMeta "language" meta `mplus` lookupMeta "lang" meta creators = getCreator "creator" meta @@ -799,6 +800,16 @@ replaceRefs refTable = walk replaceOneRef 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 -- cgit v1.2.3 From bd96f2bdcce55bfe5a7b753d29b3f4a45986f211 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 3 Dec 2013 10:32:03 -0800 Subject: EPUB writer: Allow 'stylesheet' in metadata. The value is a path to the stylesheet. --- README | 2 ++ src/Text/Pandoc/Writers/EPUB.hs | 33 ++++++++++++++++++++++----------- 2 files changed, 24 insertions(+), 11 deletions(-) (limited to 'src/Text') diff --git a/README b/README index 2a5ecc6ad..11cbf189f 100644 --- a/README +++ b/README @@ -2822,6 +2822,8 @@ The following fields are recognized: ~ A string value. `cover-image` ~ A string value (path to cover image). +`stylesheet` + ~ A string value (path to CSS stylesheet). Literate Haskell support ======================== diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 59c90b16d..32b0c3c32 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -38,7 +38,7 @@ 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 ((<$>)) @@ -89,8 +89,13 @@ data EPUBMetadata = EPUBMetadata{ , 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 @@ -262,6 +267,7 @@ metadataFromMeta opts meta = EPUBMetadata{ , epubCoverage = coverage , epubRights = rights , epubCoverImage = coverImage + , epubStylesheet = stylesheet } where identifiers = getIdentifier meta titles = getTitle meta @@ -282,6 +288,10 @@ metadataFromMeta opts meta = EPUBMetadata{ rights = metaValueToString <$> lookupMeta "rights" meta coverImage = lookup "epub-cover-image" (writerVariables opts) `mplus` (metaValueToString <$> lookupMeta "cover-image" meta) + stylesheet = (StylesheetContents <$> + lookup "epub-stylesheet" (writerVariables opts)) `mplus` + ((StylesheetPath . metaValueToString) <$> + lookupMeta "stylesheet" meta) -- | Produce an EPUB file from a Pandoc document. writeEPUB :: WriterOptions -- ^ Writer options @@ -437,7 +447,7 @@ writeEPUB opts doc@(Pandoc meta _) = do (x:_) -> identifierText x -- use first identifier as UUID [] -> error "epubIdentifier is null" -- shouldn't happen currentTime <- getCurrentTime - let contentsData = fromStringLazy $ ppTopElement $ + let contentsData = UTF8.fromStringLazy $ ppTopElement $ unode "package" ! [("version", case version of EPUB2 -> "2.0" EPUB3 -> "3.0") @@ -525,7 +535,7 @@ 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" $ @@ -557,7 +567,7 @@ 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" $ @@ -571,10 +581,10 @@ 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" $ @@ -583,18 +593,19 @@ 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 -- cgit v1.2.3 From fdaeec0c48d742489ddf0ec0c0261ca9c53f989b Mon Sep 17 00:00:00 2001 From: Jose Luis Duran Date: Mon, 2 Dec 2013 09:55:58 +0000 Subject: Add booktabs package for LaTeX tables [ci skip] --- README | 2 +- data/templates | 2 +- src/Text/Pandoc/Writers/LaTeX.hs | 10 ++--- tests/tables.latex | 90 ++++++++++++++++++++-------------------- 4 files changed, 52 insertions(+), 52 deletions(-) (limited to 'src/Text') diff --git a/README b/README index 2a5ecc6ad..3ba1e364e 100644 --- a/README +++ b/README @@ -108,7 +108,7 @@ to PDF: Production of a PDF requires that a LaTeX engine be installed (see `--latex-engine`, below), and assumes that the following LaTeX packages are available: `amssymb`, `amsmath`, `ifxetex`, `ifluatex`, `listings` (if the -`--listings` option is used), `fancyvrb`, `longtable`, `url`, +`--listings` option is used), `fancyvrb`, `longtable`, `booktabs`, `url`, `graphicx`, `hyperref`, `ulem`, `babel` (if the `lang` variable is set), `fontspec` (if `xelatex` or `lualatex` is used as the LaTeX engine), `xltxtra` and `xunicode` (if `xelatex` is used). diff --git a/data/templates b/data/templates index 713a8f63d..f643a076d 160000 --- a/data/templates +++ b/data/templates @@ -1 +1 @@ -Subproject commit 713a8f63d5589ab9313869e47b03cf7f49e00e98 +Subproject commit f643a076d8c2b0b21391fd6aa1dedb2dd84c7e63 diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index f3cbcf19f..a2e0b016f 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -453,12 +453,12 @@ 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 @@ -466,10 +466,10 @@ 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}" @@ -506,7 +506,7 @@ tableRowToLaTeX header aligns widths cols = do (w * scaleFactor))) <> (halign a <> cr <> c <> cr) <> "\\end{minipage}" let cells = zipWith3 toCell widths aligns renderedCells - return $ hsep (intersperse "&" cells) $$ "\\\\\\noalign{\\medskip}" + return $ hsep (intersperse "&" cells) $$ "\\\\\\addlinespace" listItemToLaTeX :: [Block] -> State WriterState Doc listItemToLaTeX lst = blockListToLaTeX lst >>= return . (text "\\item" $$) . diff --git a/tests/tables.latex b/tests/tables.latex index c27e10461..1a87c4f71 100644 --- a/tests/tables.latex +++ b/tests/tables.latex @@ -1,59 +1,59 @@ Simple table with caption: \begin{longtable}[c]{@{}rlcl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace Right & Left & Center & Default -\\\noalign{\medskip} -\hline\noalign{\medskip} +\\\addlinespace +\midrule\endhead 12 & 12 & 12 & 12 -\\\noalign{\medskip} +\\\addlinespace 123 & 123 & 123 & 123 -\\\noalign{\medskip} +\\\addlinespace 1 & 1 & 1 & 1 -\\\noalign{\medskip} -\hline -\noalign{\medskip} +\\\addlinespace +\bottomrule +\addlinespace \caption{Demonstration of simple table syntax.} \end{longtable} Simple table without caption: \begin{longtable}[c]{@{}rlcl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace Right & Left & Center & Default -\\\noalign{\medskip} -\hline\noalign{\medskip} +\\\addlinespace +\midrule\endhead 12 & 12 & 12 & 12 -\\\noalign{\medskip} +\\\addlinespace 123 & 123 & 123 & 123 -\\\noalign{\medskip} +\\\addlinespace 1 & 1 & 1 & 1 -\\\noalign{\medskip} -\hline +\\\addlinespace +\bottomrule \end{longtable} Simple table indented two spaces: \begin{longtable}[c]{@{}rlcl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace Right & Left & Center & Default -\\\noalign{\medskip} -\hline\noalign{\medskip} +\\\addlinespace +\midrule\endhead 12 & 12 & 12 & 12 -\\\noalign{\medskip} +\\\addlinespace 123 & 123 & 123 & 123 -\\\noalign{\medskip} +\\\addlinespace 1 & 1 & 1 & 1 -\\\noalign{\medskip} -\hline -\noalign{\medskip} +\\\addlinespace +\bottomrule +\addlinespace \caption{Demonstration of simple table syntax.} \end{longtable} Multiline table with caption: \begin{longtable}[c]{@{}clrl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace \begin{minipage}[b]{0.13\columnwidth}\centering Centered Header \end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright @@ -63,8 +63,8 @@ Right Aligned \end{minipage} & \begin{minipage}[b]{0.30\columnwidth}\raggedright Default aligned \end{minipage} -\\\noalign{\medskip} -\hline\noalign{\medskip} +\\\addlinespace +\midrule\endhead \begin{minipage}[t]{0.13\columnwidth}\centering First \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -74,7 +74,7 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Example of a row that spans multiple lines. \end{minipage} -\\\noalign{\medskip} +\\\addlinespace \begin{minipage}[t]{0.13\columnwidth}\centering Second \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -84,16 +84,16 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Here's another one. Note the blank line between rows. \end{minipage} -\\\noalign{\medskip} -\hline -\noalign{\medskip} +\\\addlinespace +\bottomrule +\addlinespace \caption{Here's the caption. It may span multiple lines.} \end{longtable} Multiline table without caption: \begin{longtable}[c]{@{}clrl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace \begin{minipage}[b]{0.13\columnwidth}\centering Centered Header \end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright @@ -103,8 +103,8 @@ Right Aligned \end{minipage} & \begin{minipage}[b]{0.30\columnwidth}\raggedright Default aligned \end{minipage} -\\\noalign{\medskip} -\hline\noalign{\medskip} +\\\addlinespace +\midrule\endhead \begin{minipage}[t]{0.13\columnwidth}\centering First \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -114,7 +114,7 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Example of a row that spans multiple lines. \end{minipage} -\\\noalign{\medskip} +\\\addlinespace \begin{minipage}[t]{0.13\columnwidth}\centering Second \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -124,27 +124,27 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Here's another one. Note the blank line between rows. \end{minipage} -\\\noalign{\medskip} -\hline +\\\addlinespace +\bottomrule \end{longtable} Table without column headers: \begin{longtable}[c]{@{}rlcr@{}} -\hline\noalign{\medskip} +\toprule\addlinespace 12 & 12 & 12 & 12 -\\\noalign{\medskip} +\\\addlinespace 123 & 123 & 123 & 123 -\\\noalign{\medskip} +\\\addlinespace 1 & 1 & 1 & 1 -\\\noalign{\medskip} -\hline +\\\addlinespace +\bottomrule \end{longtable} Multiline table without column headers: \begin{longtable}[c]{@{}clrl@{}} -\hline\noalign{\medskip} +\toprule\addlinespace \begin{minipage}[t]{0.13\columnwidth}\centering First \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -154,7 +154,7 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Example of a row that spans multiple lines. \end{minipage} -\\\noalign{\medskip} +\\\addlinespace \begin{minipage}[t]{0.13\columnwidth}\centering Second \end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright @@ -164,6 +164,6 @@ row \end{minipage} & \begin{minipage}[t]{0.30\columnwidth}\raggedright Here's another one. Note the blank line between rows. \end{minipage} -\\\noalign{\medskip} -\hline +\\\addlinespace +\bottomrule \end{longtable} -- cgit v1.2.3 From 5314df51f368853f7ee4bf8f75eedf43afc0400e Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Wed, 4 Dec 2013 10:00:40 -0800 Subject: Stop parsing "list lines" when we hit a block tag. This fixes exponential slowdown in certain input, e.g. a series of lists followed by `
`. --- src/Text/Pandoc/Readers/Markdown.hs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 33d1a9620..4cb75d86c 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -732,7 +732,8 @@ listLine = try $ do notFollowedBy' (do indentSpaces many (spaceChar) listStart) - chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 anyChar) newline + chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 (satisfy (/='<')) + <|> (notFollowedBy' (htmlTag isBlockTag) >> count 1 anyChar)) newline return $ concat chunks -- parse raw text for one list item, excluding start marker and continuations @@ -759,6 +760,7 @@ listContinuationLine :: MarkdownParser String listContinuationLine = try $ do notFollowedBy blankline notFollowedBy' listStart + notFollowedBy' $ try $ skipMany spaceChar >> htmlTag (~== TagClose "div") optional indentSpaces result <- anyLine return $ result ++ "\n" -- cgit v1.2.3 From def05d3504bfc90c31c1621438f7a51c95079ad2 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 6 Dec 2013 16:43:59 -0800 Subject: HTML reader: Parse LaTeX math if appropriate options are set. * Moved inlineMath, displayMath from Markdown reader to Parsing. * Export them from Parsing. (API change.) * Generalize their types. --- src/Text/Pandoc/Parsing.hs | 35 +++++++++++++++++++++++++++++++++++ src/Text/Pandoc/Readers/HTML.hs | 9 ++++++++- src/Text/Pandoc/Readers/LaTeX.hs | 3 ++- src/Text/Pandoc/Readers/Markdown.hs | 33 --------------------------------- 4 files changed, 45 insertions(+), 35 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Parsing.hs b/src/Text/Pandoc/Parsing.hs index 9687d7712..e15854333 100644 --- a/src/Text/Pandoc/Parsing.hs +++ b/src/Text/Pandoc/Parsing.hs @@ -48,6 +48,8 @@ module Text.Pandoc.Parsing ( (>>~), romanNumeral, emailAddress, uri, + mathInline, + mathDisplay, withHorizDisplacement, withRaw, escaped, @@ -455,6 +457,39 @@ uri = try $ do let uri' = scheme ++ ":" ++ fromEntities str' return (uri', escapeURI uri') +mathInlineWith :: String -> String -> Parser [Char] st String +mathInlineWith op cl = try $ do + string op + notFollowedBy space + words' <- many1Till (count 1 (noneOf "\n\\") + <|> (char '\\' >> anyChar >>= \c -> return ['\\',c]) + <|> count 1 newline <* notFollowedBy' blankline + *> return " ") + (try $ string cl) + notFollowedBy digit -- to prevent capture of $5 + return $ concat words' + +mathDisplayWith :: String -> String -> Parser [Char] st String +mathDisplayWith op cl = try $ do + string op + many1Till (noneOf "\n" <|> (newline >>~ notFollowedBy' blankline)) (try $ string cl) + +mathDisplay :: Parser [Char] ParserState String +mathDisplay = + (guardEnabled Ext_tex_math_dollars >> mathDisplayWith "$$" "$$") + <|> (guardEnabled Ext_tex_math_single_backslash >> + mathDisplayWith "\\[" "\\]") + <|> (guardEnabled Ext_tex_math_double_backslash >> + mathDisplayWith "\\\\[" "\\\\]") + +mathInline :: Parser [Char] ParserState String +mathInline = + (guardEnabled Ext_tex_math_dollars >> mathInlineWith "$" "$") + <|> (guardEnabled Ext_tex_math_single_backslash >> + mathInlineWith "\\(" "\\)") + <|> (guardEnabled Ext_tex_math_double_backslash >> + mathInlineWith "\\\\(" "\\\\)") + -- | Applies a parser, returns tuple of its results and its horizontal -- displacement (the difference between the source column at the end -- and the source column at the beginning). Vertical displacement diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index d691c9878..e758f712f 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -467,7 +467,13 @@ pBlank = try $ do pTagContents :: Parser [Char] ParserState Inline pTagContents = - pStr <|> pSpace <|> smartPunctuation pTagContents <|> pSymbol <|> pBad + Math InlineMath `fmap` mathInline + <|> Math DisplayMath `fmap` mathDisplay + <|> pStr + <|> pSpace + <|> smartPunctuation pTagContents + <|> pSymbol + <|> pBad pStr :: Parser [Char] ParserState Inline pStr = do @@ -482,6 +488,7 @@ isSpecial '"' = True isSpecial '\'' = True isSpecial '.' = True isSpecial '-' = True +isSpecial '$' = True isSpecial '\8216' = True isSpecial '\8217' = True isSpecial '\8220' = True diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 75e29ebb9..509cb5d74 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -38,7 +38,8 @@ import Text.Pandoc.Definition import Text.Pandoc.Walk import Text.Pandoc.Shared import Text.Pandoc.Options -import Text.Pandoc.Parsing hiding ((<|>), many, optional, space) +import Text.Pandoc.Parsing hiding ((<|>), many, optional, space, + mathDisplay, mathInline) import qualified Text.Pandoc.UTF8 as UTF8 import Data.Char ( chr, ord ) import Control.Monad diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 4cb75d86c..11168bc09 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1408,39 +1408,6 @@ math :: MarkdownParser (F Inlines) math = (return . B.displayMath <$> (mathDisplay >>= applyMacros')) <|> (return . B.math <$> (mathInline >>= applyMacros')) -mathDisplay :: MarkdownParser String -mathDisplay = - (guardEnabled Ext_tex_math_dollars >> mathDisplayWith "$$" "$$") - <|> (guardEnabled Ext_tex_math_single_backslash >> - mathDisplayWith "\\[" "\\]") - <|> (guardEnabled Ext_tex_math_double_backslash >> - mathDisplayWith "\\\\[" "\\\\]") - -mathDisplayWith :: String -> String -> MarkdownParser String -mathDisplayWith op cl = try $ do - string op - many1Till (noneOf "\n" <|> (newline >>~ notFollowedBy' blankline)) (try $ string cl) - -mathInline :: MarkdownParser String -mathInline = - (guardEnabled Ext_tex_math_dollars >> mathInlineWith "$" "$") - <|> (guardEnabled Ext_tex_math_single_backslash >> - mathInlineWith "\\(" "\\)") - <|> (guardEnabled Ext_tex_math_double_backslash >> - mathInlineWith "\\\\(" "\\\\)") - -mathInlineWith :: String -> String -> MarkdownParser String -mathInlineWith op cl = try $ do - string op - notFollowedBy space - words' <- many1Till (count 1 (noneOf "\n\\") - <|> (char '\\' >> anyChar >>= \c -> return ['\\',c]) - <|> count 1 newline <* notFollowedBy' blankline - *> return " ") - (try $ string cl) - notFollowedBy digit -- to prevent capture of $5 - return $ concat words' - -- Parses material enclosed in *s, **s, _s, or __s. -- Designed to avoid backtracking. enclosure :: Char -- cgit v1.2.3 From 4a1446705575ade1cf8ecaf19dec877bdd0ffb93 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 6 Dec 2013 17:31:47 -0800 Subject: Text.Pandoc: Don't default to pandocExtensions for all writers. In particular, we don't want to default to math parsing for the HTML writer. --- src/Text/Pandoc.hs | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc.hs b/src/Text/Pandoc.hs index 703bb876a..3ae81db00 100644 --- a/src/Text/Pandoc.hs +++ b/src/Text/Pandoc.hs @@ -267,7 +267,10 @@ getDefaultExtensions "markdown_strict" = strictExtensions getDefaultExtensions "markdown_phpextra" = phpMarkdownExtraExtensions getDefaultExtensions "markdown_mmd" = multimarkdownExtensions getDefaultExtensions "markdown_github" = githubMarkdownExtensions -getDefaultExtensions _ = pandocExtensions +getDefaultExtensions "markdown" = pandocExtensions +getDefaultExtensions "plain" = pandocExtensions +getDefaultExtensions "textile" = Set.fromList [Ext_auto_identifiers, Ext_raw_tex] +getDefaultExtensions _ = Set.fromList [Ext_auto_identifiers] -- | Retrieve reader based on formatSpec (format+extensions). getReader :: String -> Either String (ReaderOptions -> String -> IO Pandoc) -- cgit v1.2.3 From e5a7c31a32b2b53ef5073355b70dc17ecf1d16af Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 7 Dec 2013 17:12:52 -0800 Subject: Markdown reader: Fixed bug with literal `` in lists. Closes #1078. --- src/Text/Pandoc/Readers/Markdown.hs | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 11168bc09..c32c5ed86 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -732,8 +732,9 @@ listLine = try $ do notFollowedBy' (do indentSpaces many (spaceChar) listStart) - chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 (satisfy (/='<')) - <|> (notFollowedBy' (htmlTag isBlockTag) >> count 1 anyChar)) newline + notFollowedBy' $ htmlTag (~== TagClose "div") + chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 anyChar) + newline return $ concat chunks -- parse raw text for one list item, excluding start marker and continuations @@ -760,7 +761,7 @@ listContinuationLine :: MarkdownParser String listContinuationLine = try $ do notFollowedBy blankline notFollowedBy' listStart - notFollowedBy' $ try $ skipMany spaceChar >> htmlTag (~== TagClose "div") + notFollowedBy' $ htmlTag (~== TagClose "div") optional indentSpaces result <- anyLine return $ result ++ "\n" -- cgit v1.2.3 From e2c4156c20f113b6f965ccd9313c5aa062cc198f Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 7 Dec 2013 19:41:42 -0800 Subject: Small performance improvement in list parsing. --- src/Text/Pandoc/Readers/Markdown.hs | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index c32c5ed86..a948d5ad3 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -733,8 +733,9 @@ listLine = try $ do many (spaceChar) listStart) notFollowedBy' $ htmlTag (~== TagClose "div") - chunks <- manyTill (liftM snd (htmlTag isCommentTag) <|> count 1 anyChar) - newline + chunks <- manyTill (liftM snd (htmlTag isCommentTag) + <|> many1 (satisfy (/='\n')) + <|> count 1 anyChar) newline return $ concat chunks -- parse raw text for one list item, excluding start marker and continuations -- cgit v1.2.3 From 8e255fad98baf2448d8c1866d7c03c5d110a505f Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sat, 7 Dec 2013 19:56:54 -0800 Subject: Another small performance improvement. --- src/Text/Pandoc/Readers/Markdown.hs | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index a948d5ad3..e77dda8d7 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -733,9 +733,11 @@ listLine = try $ do many (spaceChar) listStart) notFollowedBy' $ htmlTag (~== TagClose "div") - chunks <- manyTill (liftM snd (htmlTag isCommentTag) - <|> many1 (satisfy (/='\n')) - <|> count 1 anyChar) newline + chunks <- manyTill + ( many1 (satisfy $ \c -> c /= '\n' && c /= '<') + <|> liftM snd (htmlTag isCommentTag) + <|> count 1 anyChar + ) newline return $ concat chunks -- parse raw text for one list item, excluding start marker and continuations -- cgit v1.2.3 From f9662957704ebfe83d7764dd64151d37c476c0b0 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Mon, 9 Dec 2013 20:31:47 -0800 Subject: Don't use tilde code blocks with braced attributes in gfm output. A consequence of this change is that the backtick form will be preferred in general if both are enabled. I think that is good, as it is much more widespread than the tilde form. Closes #1084. --- src/Text/Pandoc/Writers/Markdown.hs | 18 +++++++++--------- tests/lhs-test.markdown | 4 ++-- 2 files changed, 11 insertions(+), 11 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Markdown.hs b/src/Text/Pandoc/Writers/Markdown.hs index 60d474263..c0b189b75 100644 --- a/src/Text/Pandoc/Writers/Markdown.hs +++ b/src/Text/Pandoc/Writers/Markdown.hs @@ -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 diff --git a/tests/lhs-test.markdown b/tests/lhs-test.markdown index 47ec920d3..75a253bf4 100644 --- a/tests/lhs-test.markdown +++ b/tests/lhs-test.markdown @@ -4,11 +4,11 @@ lhs test `unsplit` is an arrow that takes a pair of values and combines them to return a single value: -~~~~ {.sourceCode .literate .haskell} +``` {.sourceCode .literate .haskell} unsplit :: (Arrow a) => (b -> c -> d) -> a (b, c) d unsplit = arr . uncurry -- arr (\op (x,y) -> x `op` y) -~~~~ +``` `(***)` combines two arrows into a new arrow by running the two arrows on a pair of values (one arrow on the first item of the pair and one arrow on the -- cgit v1.2.3 From 142f81889b6ede73f965afa01ac67427e1335e9d Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Mon, 9 Dec 2013 22:35:22 -0800 Subject: Added `withSocketsDo` around http conduit code in `openURL`. This should address #1080, but further testing on Windows is needed before we can close the bug. --- src/Text/Pandoc/Shared.hs | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index 7592b7659..ce71881e6 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -121,6 +121,7 @@ import Data.ByteString.Lazy (toChunks) import Network.HTTP.Conduit (httpLbs, parseUrl, withManager, responseBody, responseHeaders) import Network.HTTP.Types.Header ( hContentType) +import Network (withSocketsDo) #else import Network.URI (parseURI) import Network.HTTP (findHeader, rspBody, @@ -644,7 +645,7 @@ openURL u contents = B8.pack $ unEscapeString $ drop 1 $ dropWhile (/=',') u in return $ Right (decodeLenient contents, Just mime) #ifdef HTTP_CONDUIT - | otherwise = E.try $ do + | otherwise = withSocketsDo $ E.try $ do req <- parseUrl u resp <- withManager $ httpLbs req return (BS.concat $ toChunks $ responseBody resp, -- cgit v1.2.3 From 6d0cd9203ce968b96ffa8fc2fbae5a50c99b125b Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 12 Dec 2013 22:34:13 -0800 Subject: Markdown reader: Fixed regression in title blocks. If author field was empty, date was being ignored. Closes #1089. --- src/Text/Pandoc/Readers/Markdown.hs | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index e77dda8d7..8014d480c 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -215,10 +215,10 @@ pandocTitleBlock = try $ do author' <- author date' <- date return $ - ( if B.isNull title' then id else B.setMeta "title" title' - . if null author' then id else B.setMeta "author" author' - . if B.isNull date' then id else B.setMeta "date" date' ) - nullMeta + (if B.isNull title' then id else B.setMeta "title" title') + . (if null author' then id else B.setMeta "author" author') + . (if B.isNull date' then id else B.setMeta "date" date') + $ nullMeta updateState $ \st -> st{ stateMeta' = stateMeta' st <> meta' } yamlMetaBlock :: MarkdownParser (F Blocks) -- cgit v1.2.3 From ca3c292f30c04bd24287d554f08a7911c808a4e2 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Fri, 13 Dec 2013 11:10:04 -0800 Subject: EPUB writer: Fixed bug with `--epub-stylesheet`. Now the contents of `writerEpubStylesheet` (set by `--epub-stylesheet`) should again work, and take precedence over a stylesheet specified in the metadata. --- src/Text/Pandoc/Writers/EPUB.hs | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 32b0c3c32..36ead0b8f 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -288,8 +288,7 @@ metadataFromMeta opts meta = EPUBMetadata{ rights = metaValueToString <$> lookupMeta "rights" meta coverImage = lookup "epub-cover-image" (writerVariables opts) `mplus` (metaValueToString <$> lookupMeta "cover-image" meta) - stylesheet = (StylesheetContents <$> - lookup "epub-stylesheet" (writerVariables opts)) `mplus` + stylesheet = (StylesheetContents <$> writerEpubStylesheet opts) `mplus` ((StylesheetPath . metaValueToString) <$> lookupMeta "stylesheet" meta) -- cgit v1.2.3 From 5adbe7b36592ba8779563bba791ad4724fde927c Mon Sep 17 00:00:00 2001 From: Jeff Arnold Date: Fri, 13 Dec 2013 19:16:04 -0500 Subject: LaTeX reader: add support for Verb macro --- src/Text/Pandoc/Readers/LaTeX.hs | 1 + 1 file changed, 1 insertion(+) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 509cb5d74..736f2ff1c 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -455,6 +455,7 @@ inlineCommands = M.fromList $ , ("footnote", (note . mconcat) <$> (char '{' *> manyTill block (char '}'))) , ("verb", doverb) , ("lstinline", doverb) + , ("Verb", doverb) , ("texttt", (code . stringify . toList) <$> tok) , ("url", (unescapeURL <$> braced) >>= \url -> pure (link url "" (str url))) -- cgit v1.2.3 From 2f00f5c7c22ef879df3a21f1c3da967d106628e9 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Sun, 15 Dec 2013 12:27:29 -0800 Subject: Properly handle script blocks in strict mode. (That is, markdown-markdown_in_html_blocks.) Previously a spurious `

` tag was being added. Closes #1093. --- src/Text/Pandoc/Readers/Markdown.hs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 8014d480c..b2e88d47e 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -902,7 +902,9 @@ plain = fmap B.plain . trimInlinesF . mconcat <$> many1 inline -- htmlElement :: MarkdownParser String -htmlElement = strictHtmlBlock <|> liftM snd (htmlTag isBlockTag) +htmlElement = rawVerbatimBlock + <|> strictHtmlBlock + <|> liftM snd (htmlTag isBlockTag) htmlBlock :: MarkdownParser (F Blocks) htmlBlock = do -- cgit v1.2.3 From 826443926f85b72840b49cd7973e91ed80a09b5d Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Mon, 16 Dec 2013 13:58:54 -0800 Subject: Docbook reader: Avoid failure if tbody contains no tr or row elements. --- src/Text/Pandoc/Readers/DocBook.hs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/DocBook.hs b/src/Text/Pandoc/Readers/DocBook.hs index fc29988d5..03c6140ac 100644 --- a/src/Text/Pandoc/Readers/DocBook.hs +++ b/src/Text/Pandoc/Readers/DocBook.hs @@ -806,7 +806,9 @@ parseBlock (Elem e) = (x >= '0' && x <= '9') || x == '.') w Nothing -> 0 :: Double - let numrows = maximum $ map length bodyrows + let numrows = case bodyrows of + [] -> 0 + xs -> maximum $ map length xs let aligns = case colspecs of [] -> replicate numrows AlignDefault cs -> map toAlignment cs -- cgit v1.2.3 From 0132f6fcb75e58774846e92bac6df2afd7a50e6f Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 17 Dec 2013 16:03:43 -0800 Subject: LaTeX reader: Support babel-style quoting: `` "`..."' ``. --- src/Text/Pandoc/Readers/LaTeX.hs | 2 ++ 1 file changed, 2 insertions(+) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 736f2ff1c..0020a8f26 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -173,6 +173,8 @@ double_quote :: LP Inlines double_quote = ( quoted' doubleQuoted (try $ string "``") (void $ try $ string "''") <|> quoted' doubleQuoted (string "“") (void $ char '”') + -- the following is used by babel for localized quotes: + <|> quoted' doubleQuoted (try $ string "\"`") (void $ try $ string "\"'") <|> quoted' doubleQuoted (string "\"") (void $ char '"') ) -- cgit v1.2.3 From a3f6f2827c9acb0e68ac533745b3f4dc8836ee24 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 17 Dec 2013 20:10:09 -0800 Subject: LaTeX writer: Factored out function for table cell creation. --- src/Text/Pandoc/Writers/LaTeX.hs | 30 +++++++++++++++++------------- 1 file changed, 17 insertions(+), 13 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index a2e0b016f..b7beacd91 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -490,24 +490,28 @@ 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 + let widths' = map (scaleFactor *) widths + cells <- mapM (tableCellToLaTeX header) $ zip3 widths' aligns cols return $ hsep (intersperse "&" cells) $$ "\\\\\\addlinespace" +tableCellToLaTeX :: Bool -> (Double, Alignment, [Block]) + -> State WriterState Doc +tableCellToLaTeX _ (0, _, blocks) = blockListToLaTeX blocks +tableCellToLaTeX header (width, align, blocks) = do + cellContents <- blockListToLaTeX blocks + let valign = text $ if header then "[b]" else "[t]" + let halign = case align of + AlignLeft -> "\\raggedright" + AlignRight -> "\\raggedleft" + AlignCenter -> "\\centering" + AlignDefault -> "\\raggedright" + return $ "\\begin{minipage}" <> valign <> + braces (text (printf "%.2f\\columnwidth" width)) <> + (halign <> cr <> cellContents <> cr) <> "\\end{minipage}" + listItemToLaTeX :: [Block] -> State WriterState Doc listItemToLaTeX lst = blockListToLaTeX lst >>= return . (text "\\item" $$) . (nest 2) -- cgit v1.2.3 From 8053ba2123699fd89900b2a9c5656a72d3a2fe85 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Tue, 17 Dec 2013 20:53:59 -0800 Subject: LaTeX writer: Better treatment of footnotes in tables. Notes now appear in the regular sequence, rather than in the table cell. (This was a regression in 1.10.) --- src/Text/Pandoc/Writers/LaTeX.hs | 35 +++++++++++++++++++++++++++++------ 1 file changed, 29 insertions(+), 6 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index b7beacd91..7612be100 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -50,6 +50,8 @@ 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 +72,7 @@ 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, @@ -501,16 +503,31 @@ 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}" + 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" $$) . @@ -713,14 +730,20 @@ 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:[]) -- cgit v1.2.3 From 1ed2c467c9442934c060257f6e191a5a3d6c1e38 Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 17:06:27 -0500 Subject: HLint: Use all Replace `and . map` with `all`. --- src/Text/Pandoc/Shared.hs | 2 +- src/Text/Pandoc/Writers/LaTeX.hs | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index ce71881e6..cd2f7e24d 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -532,7 +532,7 @@ headerShift n = walk shift -- | Detect if a list is tight. isTightList :: [[Block]] -> Bool -isTightList = and . map firstIsPlain +isTightList = all firstIsPlain where firstIsPlain (Plain _ : _) = True firstIsPlain _ = False diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index 7612be100..47b769c48 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -443,7 +443,7 @@ 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 $$ vcat items $$ @@ -764,9 +764,9 @@ 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 -- cgit v1.2.3 From 0c5e7cf8cb4fe6959d7e89880e8925afe6625414 Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 20:19:24 -0500 Subject: HLint: use `elem` and `notElem` Replaces long conditional chains with calls to `elem` and `notElem`. --- pandoc.hs | 2 +- src/Text/Pandoc/Parsing.hs | 4 ++-- src/Text/Pandoc/Readers/LaTeX.hs | 5 ++--- src/Text/Pandoc/Readers/Markdown.hs | 8 ++++---- src/Text/Pandoc/Shared.hs | 12 ++++-------- src/Text/Pandoc/Writers/Markdown.hs | 11 ++++++----- src/Text/Pandoc/Writers/Org.hs | 2 +- 7 files changed, 20 insertions(+), 24 deletions(-) (limited to 'src/Text') diff --git a/pandoc.hs b/pandoc.hs index cada3347d..ccd3e57fb 100644 --- a/pandoc.hs +++ b/pandoc.hs @@ -656,7 +656,7 @@ options = (ReqArg (\arg opt -> do let b = takeBaseName arg - if (b == "pdflatex" || b == "lualatex" || b == "xelatex") + if b `elem` ["pdflatex", "lualatex", "xelatex"] then return opt { optLaTeXEngine = arg } else err 45 "latex-engine must be pdflatex, lualatex, or xelatex.") "PROGRAM") diff --git a/src/Text/Pandoc/Parsing.hs b/src/Text/Pandoc/Parsing.hs index e15854333..2f21e1253 100644 --- a/src/Text/Pandoc/Parsing.hs +++ b/src/Text/Pandoc/Parsing.hs @@ -271,7 +271,7 @@ spaceChar = satisfy $ \c -> c == ' ' || c == '\t' -- | Parses a nonspace, nonnewline character. nonspaceChar :: Parser [Char] st Char -nonspaceChar = satisfy $ \x -> x /= '\t' && x /= '\n' && x /= ' ' && x /= '\r' +nonspaceChar = satisfy $ flip notElem ['\t', '\n', ' ', '\r'] -- | Skips zero or more spaces or tabs. skipSpaces :: Parser [Char] st () @@ -1062,7 +1062,7 @@ doubleQuoteStart :: Parser [Char] ParserState () doubleQuoteStart = do failIfInQuoteContext InDoubleQuote try $ do charOrRef "\"\8220\147" - notFollowedBy (satisfy (\c -> c == ' ' || c == '\t' || c == '\n')) + notFollowedBy . satisfy $ flip elem [' ', '\t', '\n'] doubleQuoteEnd :: Parser [Char] st () doubleQuoteEnd = do diff --git a/src/Text/Pandoc/Readers/LaTeX.hs b/src/Text/Pandoc/Readers/LaTeX.hs index 0020a8f26..51271edc5 100644 --- a/src/Text/Pandoc/Readers/LaTeX.hs +++ b/src/Text/Pandoc/Readers/LaTeX.hs @@ -874,9 +874,8 @@ verbatimEnv = do (_,r) <- withRaw $ do controlSeq "begin" name <- braced - guard $ name == "verbatim" || name == "Verbatim" || - name == "lstlisting" || name == "minted" || - name == "alltt" + guard $ name `elem` ["verbatim", "Verbatim", "lstlisting", + "minted", "alltt"] verbEnv name rest <- getInput return (r,rest) diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index b2e88d47e..f483ab059 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -789,8 +789,8 @@ listItem start = try $ do orderedList :: MarkdownParser (F Blocks) orderedList = try $ do (start, style, delim) <- lookAhead anyOrderedListStart - unless ((style == DefaultStyle || style == Decimal || style == Example) && - (delim == DefaultDelim || delim == Period)) $ + unless (style `elem` [DefaultStyle, Decimal, Example] && + delim `elem` [DefaultDelim, Period]) $ guardEnabled Ext_fancy_lists when (style == Example) $ guardEnabled Ext_example_lists items <- fmap sequence $ many1 $ listItem @@ -925,8 +925,8 @@ strictHtmlBlock = htmlInBalanced (not . isInlineTag) rawVerbatimBlock :: MarkdownParser String rawVerbatimBlock = try $ do - (TagOpen tag _, open) <- htmlTag (tagOpen (\t -> - t == "pre" || t == "style" || t == "script") + (TagOpen tag _, open) <- htmlTag (tagOpen (flip elem + ["pre", "style", "script"]) (const True)) contents <- manyTill anyChar (htmlTag (~== TagClose tag)) return $ open ++ contents ++ renderTags [TagClose tag] diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index cd2f7e24d..3446f4343 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -564,14 +564,10 @@ makeMeta title authors date = -- | Render HTML tags. renderTags' :: [Tag String] -> String renderTags' = renderTagsOptions - renderOptions{ optMinimize = \x -> - let y = map toLower x - in y == "hr" || y == "br" || - y == "img" || y == "meta" || - y == "link" - , optRawTag = \x -> - let y = map toLower x - in y == "script" || y == "style" } + renderOptions{ optMinimize = matchTags ["hr", "br", "img", + "meta", "link"] + , optRawTag = matchTags ["script", "style"] } + where matchTags = \tags -> flip elem tags . map toLower -- -- File handling diff --git a/src/Text/Pandoc/Writers/Markdown.hs b/src/Text/Pandoc/Writers/Markdown.hs index c0b189b75..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 @@ -628,10 +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/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 -- cgit v1.2.3 From c35f5ba42df094cef5f69a191315385a0a1e12b0 Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 20:28:53 -0500 Subject: HLint: Remove lambdas. --- src/Text/Pandoc/Writers/Shared.hs | 3 +-- tests/Tests/Walk.hs | 4 ++-- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Shared.hs b/src/Text/Pandoc/Writers/Shared.hs index 9cb08803c..33091ea94 100644 --- a/src/Text/Pandoc/Writers/Shared.hs +++ b/src/Text/Pandoc/Writers/Shared.hs @@ -65,8 +65,7 @@ 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 diff --git a/tests/Tests/Walk.hs b/tests/Tests/Walk.hs index f6aa1beae..34350e28a 100644 --- a/tests/Tests/Walk.hs +++ b/tests/Tests/Walk.hs @@ -21,11 +21,11 @@ tests = [ testGroup "Walk" p_walk :: (Typeable a, Walkable a Pandoc) => (a -> a) -> Pandoc -> Bool -p_walk f = (\(d :: Pandoc) -> everywhere (mkT f) d == walk f d) +p_walk f d = everywhere (mkT f) d == walk f d p_query :: (Eq a, Typeable a1, Monoid a, Walkable a1 Pandoc) => (a1 -> a) -> Pandoc -> Bool -p_query f = (\(d :: Pandoc) -> everything mappend (mempty `mkQ` f) d == query f d) +p_query f d = everything mappend (mempty `mkQ` f) d == query f d inlineTrans :: Inline -> Inline inlineTrans (Str xs) = Str $ map toUpper xs -- cgit v1.2.3 From f6d151889c8fff303be8ee8a4f9be67a04de9210 Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 20:43:25 -0500 Subject: HLint: redundant parens Remove parens enclosing a single element. --- pandoc.hs | 6 ++---- src/Text/Pandoc/Readers/Markdown.hs | 2 +- src/Text/Pandoc/Readers/Textile.hs | 2 +- src/Text/Pandoc/Writers/HTML.hs | 2 +- src/Text/Pandoc/Writers/MediaWiki.hs | 2 +- src/Text/Pandoc/Writers/Textile.hs | 2 +- tests/Tests/Readers/LaTeX.hs | 2 +- tests/Tests/Readers/Markdown.hs | 4 ++-- 8 files changed, 10 insertions(+), 12 deletions(-) (limited to 'src/Text') diff --git a/pandoc.hs b/pandoc.hs index ccd3e57fb..574c89771 100644 --- a/pandoc.hs +++ b/pandoc.hs @@ -1034,12 +1034,10 @@ main = do variables' <- case mathMethod of LaTeXMathML Nothing -> do - s <- readDataFileUTF8 datadir - ("LaTeXMathML.js") + s <- readDataFileUTF8 datadir "LaTeXMathML.js" return $ ("mathml-script", s) : variables MathML Nothing -> do - s <- readDataFileUTF8 datadir - ("MathMLinHTML.js") + s <- readDataFileUTF8 datadir "MathMLinHTML.js" return $ ("mathml-script", s) : variables _ -> return variables diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index f483ab059..166c524ef 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -730,7 +730,7 @@ listStart = bulletListStart <|> (anyOrderedListStart >> return ()) listLine :: MarkdownParser String listLine = try $ do notFollowedBy' (do indentSpaces - many (spaceChar) + many spaceChar listStart) notFollowedBy' $ htmlTag (~== TagClose "div") chunks <- manyTill diff --git a/src/Text/Pandoc/Readers/Textile.hs b/src/Text/Pandoc/Readers/Textile.hs index 23e07f621..93658cdea 100644 --- a/src/Text/Pandoc/Readers/Textile.hs +++ b/src/Text/Pandoc/Readers/Textile.hs @@ -594,7 +594,7 @@ surrounded border = enclosed (border *> notFollowedBy (oneOf " \t\n\r")) (try bo simpleInline :: Parser [Char] ParserState t -- ^ surrounding parser -> ([Inline] -> Inline) -- ^ Inline constructor -> Parser [Char] ParserState Inline -- ^ content parser (to be used repeatedly) -simpleInline border construct = surrounded border (inlineWithAttribute) >>= +simpleInline border construct = surrounded border inlineWithAttribute >>= return . construct . normalizeSpaces where inlineWithAttribute = (try $ optional attributes) >> inline diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index 641652276..2c6435457 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -524,7 +524,7 @@ 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 : 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/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 diff --git a/tests/Tests/Readers/LaTeX.hs b/tests/Tests/Readers/LaTeX.hs index c1efd1b68..8ff23ebc1 100644 --- a/tests/Tests/Readers/LaTeX.hs +++ b/tests/Tests/Readers/LaTeX.hs @@ -21,7 +21,7 @@ tests = [ testGroup "basic" [ "simple" =: "word" =?> para "word" , "space" =: - "some text" =?> para ("some text") + "some text" =?> para "some text" , "emphasized" =: "\\emph{emphasized}" =?> para (emph "emphasized") ] diff --git a/tests/Tests/Readers/Markdown.hs b/tests/Tests/Readers/Markdown.hs index b04ff9a0d..492680a35 100644 --- a/tests/Tests/Readers/Markdown.hs +++ b/tests/Tests/Readers/Markdown.hs @@ -171,13 +171,13 @@ tests = [ testGroup "inline code" , testGroup "smart punctuation" [ test markdownSmart "quote before ellipses" ("'...hi'" - =?> para (singleQuoted ("…hi"))) + =?> para (singleQuoted "…hi")) , test markdownSmart "apostrophe before emph" ("D'oh! A l'*aide*!" =?> para ("D’oh! A l’" <> emph "aide" <> "!")) , test markdownSmart "apostrophe in French" ("À l'arrivée de la guerre, le thème de l'«impossibilité du socialisme»" - =?> para ("À l’arrivée de la guerre, le thème de l’«impossibilité du socialisme»")) + =?> para "À l’arrivée de la guerre, le thème de l’«impossibilité du socialisme»") ] , testGroup "mixed emphasis and strong" [ "emph and strong emph alternating" =: -- cgit v1.2.3 From c8fc0a03748ceb4dbea502297ece89eac9d73aff Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 20:46:11 -0500 Subject: HLint: use /= --- src/Text/Pandoc/Writers/LaTeX.hs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index 47b769c48..1deacecb4 100644 --- a/src/Text/Pandoc/Writers/LaTeX.hs +++ b/src/Text/Pandoc/Writers/LaTeX.hs @@ -190,7 +190,7 @@ 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 } -- cgit v1.2.3 From 3d70059a4896c66cab59832a3ed9eb8334b84a2f Mon Sep 17 00:00:00 2001 From: Henry de Valence Date: Thu, 19 Dec 2013 21:07:09 -0500 Subject: HLint: use fromMaybe Replace uses of `maybe x id` with `fromMaybe x`. --- src/Text/Pandoc/PDF.hs | 3 ++- src/Text/Pandoc/Readers/DocBook.hs | 7 ++++--- src/Text/Pandoc/Readers/HTML.hs | 4 ++-- src/Text/Pandoc/Readers/Markdown.hs | 4 ++-- src/Text/Pandoc/Readers/MediaWiki.hs | 7 ++++--- src/Text/Pandoc/Writers/Docx.hs | 6 ++++-- src/Text/Pandoc/Writers/EPUB.hs | 12 ++++++------ src/Text/Pandoc/Writers/HTML.hs | 4 ++-- src/Text/Pandoc/Writers/LaTeX.hs | 3 ++- src/Text/Pandoc/Writers/ODT.hs | 3 ++- 10 files changed, 30 insertions(+), 23 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/PDF.hs b/src/Text/Pandoc/PDF.hs index e8683b98f..360338f8f 100644 --- a/src/Text/Pandoc/PDF.hs +++ b/src/Text/Pandoc/PDF.hs @@ -41,6 +41,7 @@ import System.Directory import System.Environment import Control.Monad (unless) import Data.List (isInfixOf) +import Data.Maybe (fromMaybe) import qualified Data.ByteString.Base64 as B64 import qualified Text.Pandoc.UTF8 as UTF8 import Text.Pandoc.Definition @@ -87,7 +88,7 @@ handleImage' baseURL tmpdir (Image ils (src,tit)) = do res <- fetchItem baseURL src case res of Right (contents, Just mime) -> do - let ext = maybe (takeExtension src) id $ + let ext = fromMaybe (takeExtension src) $ extensionFromMimeType mime let basename = UTF8.toString $ B64.encode $ UTF8.fromString src let fname = tmpdir basename <.> ext diff --git a/src/Text/Pandoc/Readers/DocBook.hs b/src/Text/Pandoc/Readers/DocBook.hs index 03c6140ac..56cb16b20 100644 --- a/src/Text/Pandoc/Readers/DocBook.hs +++ b/src/Text/Pandoc/Readers/DocBook.hs @@ -12,6 +12,7 @@ import Data.Char (isSpace) import Control.Monad.State import Control.Applicative ((<$>)) import Data.List (intersperse) +import Data.Maybe (fromMaybe) {- @@ -683,7 +684,7 @@ parseBlock (Elem e) = "lowerroman" -> LowerRoman "upperroman" -> UpperRoman _ -> Decimal - let start = maybe 1 id $ + let start = fromMaybe 1 $ (attrValue "override" <$> filterElement (named "listitem") e) >>= safeRead orderedListWith (start,listStyle,DefaultDelim) @@ -779,7 +780,7 @@ parseBlock (Elem e) = caption <- case filterChild isCaption e of Just t -> getInlines t Nothing -> return mempty - let e' = maybe e id $ filterChild (named "tgroup") e + let e' = fromMaybe e $ filterChild (named "tgroup") e let isColspec x = named "colspec" x || named "col" x let colspecs = case filterChild (named "colgroup") e' of Just c -> filterChildren isColspec c @@ -801,7 +802,7 @@ parseBlock (Elem e) = Just "center" -> AlignCenter _ -> AlignDefault let toWidth c = case findAttr (unqual "colwidth") c of - Just w -> maybe 0 id + Just w -> fromMaybe 0 $ safeRead $ '0': filter (\x -> (x >= '0' && x <= '9') || x == '.') w diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index e758f712f..4b44a3a21 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -207,7 +207,7 @@ pHeader = try $ do let bodyTitle = TagOpen tagtype attr ~== TagOpen "h1" [("class","title")] let level = read (drop 1 tagtype) contents <- liftM concat $ manyTill inline (pCloses tagtype <|> eof) - let ident = maybe "" id $ lookup "id" attr + let ident = fromMaybe "" $ lookup "id" attr let classes = maybe [] words $ lookup "class" attr let keyvals = [(k,v) | (k,v) <- attr, k /= "class", k /= "id"] return $ if bodyTitle @@ -257,7 +257,7 @@ pCol = try $ do skipMany pBlank return $ case lookup "width" attribs of Just x | not (null x) && last x == '%' -> - maybe 0.0 id $ safeRead ('0':'.':init x) + fromMaybe 0.0 $ safeRead ('0':'.':init x) _ -> 0.0 pColgroup :: TagParser [Double] diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 166c524ef..3a5d29b4e 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1722,7 +1722,7 @@ spanHtml = try $ do guardEnabled Ext_markdown_in_html_blocks (TagOpen _ attrs, _) <- htmlTag (~== TagOpen "span" []) contents <- mconcat <$> manyTill inline (htmlTag (~== TagClose "span")) - let ident = maybe "" id $ lookup "id" attrs + let ident = fromMaybe "" $ lookup "id" attrs let classes = maybe [] words $ lookup "class" attrs let keyvals = [(k,v) | (k,v) <- attrs, k /= "id" && k /= "class"] return $ B.spanWith (ident, classes, keyvals) <$> contents @@ -1732,7 +1732,7 @@ divHtml = try $ do guardEnabled Ext_markdown_in_html_blocks (TagOpen _ attrs, _) <- htmlTag (~== TagOpen "div" []) contents <- mconcat <$> manyTill block (htmlTag (~== TagClose "div")) - let ident = maybe "" id $ lookup "id" attrs + let ident = fromMaybe "" $ lookup "id" attrs let classes = maybe [] words $ lookup "class" attrs let keyvals = [(k,v) | (k,v) <- attrs, k /= "id" && k /= "class"] return $ B.divWith (ident, classes, keyvals) <$> contents diff --git a/src/Text/Pandoc/Readers/MediaWiki.hs b/src/Text/Pandoc/Readers/MediaWiki.hs index 1c074e3de..8d8ea0199 100644 --- a/src/Text/Pandoc/Readers/MediaWiki.hs +++ b/src/Text/Pandoc/Readers/MediaWiki.hs @@ -54,6 +54,7 @@ import Data.Sequence (viewl, ViewL(..), (<|)) import qualified Data.Foldable as F import qualified Data.Map as M import Data.Char (isDigit, isSpace) +import Data.Maybe (fromMaybe) -- | Read mediawiki from an input string and return a Pandoc document. readMediaWiki :: ReaderOptions -- ^ Reader options @@ -204,7 +205,7 @@ table = do tableStart styles <- option [] parseAttrs <* blankline let tableWidth = case lookup "width" styles of - Just w -> maybe 1.0 id $ parseWidth w + Just w -> fromMaybe 1.0 $ parseWidth w Nothing -> 1.0 caption <- option mempty tableCaption optional rowsep @@ -285,7 +286,7 @@ tableCell = try $ do Just "center" -> AlignCenter _ -> AlignDefault let width = case lookup "width" attrs of - Just xs -> maybe 0.0 id $ parseWidth xs + Just xs -> fromMaybe 0.0 $ parseWidth xs Nothing -> 0.0 return ((align, width), bs) @@ -387,7 +388,7 @@ orderedList = spaces items <- many (listItem '#' <|> li) optional (htmlTag (~== TagClose "ol")) - let start = maybe 1 id $ safeRead $ fromAttrib "start" tag + let start = fromMaybe 1 $ safeRead $ fromAttrib "start" tag return $ B.orderedListWith (start, DefaultStyle, DefaultDelim) items definitionList :: MWParser Blocks diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 5c7341b69..32ba7715a 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -29,6 +29,7 @@ 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.Maybe (fromMaybe) import Data.List ( intercalate, groupBy ) import qualified Data.ByteString as B import qualified Data.ByteString.Lazy as BL @@ -130,7 +131,8 @@ 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") @@ -322,7 +324,7 @@ 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 diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 36ead0b8f..4daa9609e 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -176,8 +176,8 @@ addMetadataFromXML e@(Element (QName name _ (Just "dc")) attrs _ _) md , titleFileAs = getAttr "file-as" , titleType = getAttr "type" } : epubTitle md } - | name == "date" = md{ epubDate = maybe "" id $ normalizeDate' - $ strContent e } + | name == "date" = md{ epubDate = fromMaybe "" $ normalizeDate' + $ strContent e } | name == "language" = md{ epubLanguage = strContent e } | name == "creator" = md{ epubCreator = Creator{ creatorText = strContent e @@ -271,7 +271,7 @@ metadataFromMeta opts meta = EPUBMetadata{ } where identifiers = getIdentifier meta titles = getTitle meta - date = maybe "" id $ + date = fromMaybe "" $ (metaValueToString <$> lookupMeta "date" meta) >>= normalizeDate' language = maybe "" metaValueToString $ lookupMeta "language" meta `mplus` lookupMeta "lang" meta @@ -297,7 +297,7 @@ 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 @@ -401,7 +401,7 @@ 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 @@ -436,7 +436,7 @@ 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)] $ () + ("media-type", fromMaybe "" $ getMimeType $ eRelativePath ent)] $ () let plainTitle = case docTitle meta of [] -> case epubTitle metadata of [] -> "UNTITLED" diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index 2c6435457..129776363 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 diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs index 1deacecb4..a76d6d82b 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 @@ -240,7 +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') diff --git a/src/Text/Pandoc/Writers/ODT.hs b/src/Text/Pandoc/Writers/ODT.hs index cc0a06243..25cd5ae13 100644 --- a/src/Text/Pandoc/Writers/ODT.hs +++ b/src/Text/Pandoc/Writers/ODT.hs @@ -31,6 +31,7 @@ Conversion of 'Pandoc' documents to ODT. module Text.Pandoc.Writers.ODT ( writeODT ) where import Data.IORef import Data.List ( isPrefixOf ) +import Data.Maybe ( fromMaybe ) import qualified Data.ByteString.Lazy as B import Text.Pandoc.UTF8 ( fromStringLazy ) import Codec.Archive.Zip @@ -127,7 +128,7 @@ 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 -- cgit v1.2.3 From d6ec6cf9cf5731977fa0f476cabef4786edd5665 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Wed, 1 Jan 2014 09:01:05 -0800 Subject: Docx writer: Fixed problem with some modified reference docx files. Include `word/_rels/settings.xml.rels` if it exists, as well as other `rels` files besides the ones pandoc generates explicitly. --- src/Text/Pandoc/Writers/Docx.hs | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 32ba7715a..67d202010 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -30,7 +30,7 @@ Conversion of 'Pandoc' documents to docx. -} module Text.Pandoc.Writers.Docx ( writeDocx ) where import Data.Maybe (fromMaybe) -import Data.List ( intercalate, groupBy ) +import Data.List ( intercalate, groupBy, isPrefixOf ) import qualified Data.ByteString as B import qualified Data.ByteString.Lazy as BL import qualified Data.ByteString.Lazy.Char8 as BL8 @@ -262,6 +262,11 @@ 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 + , f /= "word/_rels/document.xml.rels" + , f /= "word/_rels/footnotes.xml.rels" ] + miscRelEntries <- mapM entryFromArchive miscRels -- Create archive let archive = foldr addEntryToArchive emptyArchive $ @@ -269,7 +274,7 @@ 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] -- cgit v1.2.3 From f3ee82373b4ad8e955db12d3c2c2159a2bea53a0 Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Wed, 1 Jan 2014 09:22:37 -0800 Subject: HTML reader: Parse name/content pairs from meta tags as metadata. Closes #1106. --- src/Text/Pandoc/Readers/HTML.hs | 11 ++++++++++- tests/html-reader.native | 2 +- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index 4b44a3a21..506fe7770 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -76,9 +76,18 @@ pBody :: TagParser [Block] pBody = pInTags "body" block pHead :: TagParser [Block] -pHead = pInTags "head" $ pTitle <|> ([] <$ pAnyTag) +pHead = pInTags "head" $ pTitle <|> pMetaTag <|> ([] <$ pAnyTag) where pTitle = pInTags "title" inline >>= setTitle . normalizeSpaces setTitle t = [] <$ (updateState $ B.setMeta "title" (B.fromList t)) + pMetaTag = do + mt <- pSatisfy (~== TagOpen "meta" []) + let name = fromAttrib "name" mt + if null name + then return [] + else do + let content = fromAttrib "content" mt + updateState $ B.setMeta name (B.text content) + return [] block :: TagParser [Block] block = choice diff --git a/tests/html-reader.native b/tests/html-reader.native index 794512426..e80905729 100644 --- a/tests/html-reader.native +++ b/tests/html-reader.native @@ -1,4 +1,4 @@ -Pandoc (Meta {unMeta = fromList [("title",MetaInlines [Str "Pandoc",Space,Str "Test",Space,Str "Suite"])]}) +Pandoc (Meta {unMeta = fromList [("generator",MetaInlines [Str "pandoc"]),("title",MetaInlines [Str "Pandoc",Space,Str "Test",Space,Str "Suite"])]}) [Para [Str "This",Space,Str "is",Space,Str "a",Space,Str "set",Space,Str "of",Space,Str "tests",Space,Str "for",Space,Str "pandoc",Str ".",Space,Str "Most",Space,Str "of",Space,Str "them",Space,Str "are",Space,Str "adapted",Space,Str "from",Space,Str "John",Space,Str "Gruber",Str "'",Str "s",Space,Str "markdown",Space,Str "test",Space,Str "suite",Str "."] ,HorizontalRule ,Header 1 ("",[],[]) [Str "Headers"] -- cgit v1.2.3 From e3d48da6271a37129ae5bdb6cb57f006de8c5bfc Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 2 Jan 2014 15:22:50 -0800 Subject: Moved fixDisplayMath from Docx writer to Writer.Shared. --- src/Text/Pandoc/Writers/Docx.hs | 27 ++------------------------- src/Text/Pandoc/Writers/Shared.hs | 27 +++++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 25 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 67d202010..25739f7c8 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -30,7 +30,7 @@ Conversion of 'Pandoc' documents to docx. -} module Text.Pandoc.Writers.Docx ( writeDocx ) where import Data.Maybe (fromMaybe) -import Data.List ( intercalate, groupBy, isPrefixOf ) +import Data.List ( intercalate, isPrefixOf ) import qualified Data.ByteString as B import qualified Data.ByteString.Lazy as BL import qualified Data.ByteString.Lazy.Char8 as BL8 @@ -43,6 +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 ) @@ -817,27 +818,3 @@ parseXml refArchive relpath = case (findEntryByPath relpath refArchive >>= parseXMLDoc . UTF8.toStringLazy . fromEntry) of Just d -> return d 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/Shared.hs b/src/Text/Pandoc/Writers/Shared.hs index 33091ea94..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. @@ -136,3 +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 -- cgit v1.2.3 From ac100f27249e4c572e687d027330ca2cc53cb1ed Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 2 Jan 2014 15:23:16 -0800 Subject: OpenDocument writer: Fixed RawInline, RawBlock so they don't escape. --- src/Text/Pandoc/Writers/OpenDocument.hs | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/OpenDocument.hs b/src/Text/Pandoc/Writers/OpenDocument.hs index 4ddfd7166..0029c3296 100644 --- a/src/Text/Pandoc/Writers/OpenDocument.hs +++ b/src/Text/Pandoc/Writers/OpenDocument.hs @@ -285,8 +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) @@ -298,8 +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 @@ -378,8 +382,8 @@ 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" - then return $ 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 = mkImg s t -- cgit v1.2.3 From 33955fd2efd640becfddb262c4443509521cb6cc Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 2 Jan 2014 15:23:40 -0800 Subject: ODT writer: Use mathml for proper rendering of formulas. Note: LibreOffice's support for this seems a bit buggy. But it should be better than what we had before. --- src/Text/Pandoc/Writers/ODT.hs | 50 +++++++++++++++++++++++++++++++++++------- 1 file changed, 42 insertions(+), 8 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/ODT.hs b/src/Text/Pandoc/Writers/ODT.hs index 25cd5ae13..c3652d65d 100644 --- a/src/Text/Pandoc/Writers/ODT.hs +++ b/src/Text/Pandoc/Writers/ODT.hs @@ -30,8 +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 @@ -41,13 +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 @@ -61,9 +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 @@ -73,7 +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) @@ -81,6 +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 "" @@ -92,6 +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 @@ -119,8 +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 @@ -137,5 +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 -- cgit v1.2.3 From b2db6979fedc5f66d45596c9564a8313af90b9fa Mon Sep 17 00:00:00 2001 From: John MacFarlane Date: Thu, 2 Jan 2014 19:32:13 -0800 Subject: Use isHeaderBlock from Shared rather than defining it anew... --- src/Text/Pandoc/Writers/FB2.hs | 6 ++---- src/Text/Pandoc/Writers/Texinfo.hs | 6 +----- 2 files changed, 3 insertions(+), 9 deletions(-) (limited to 'src/Text') 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

and are allowed within in FB2. formatTitle :: [Inline] -> [Content] 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) = -- cgit v1.2.3 From 4e7aadb903548bb08c9132c93699071e52f61d76 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 2 Jan 2014 19:59:33 -0800 Subject: HTML writer: With --toc, headers no longer link to themselves. Closes #1081. --- src/Text/Pandoc/Writers/HTML.hs | 23 ++++++++--------------- 1 file changed, 8 insertions(+), 15 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index 129776363..c8a7e802b 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -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 -- cgit v1.2.3 From 452a140d0cd808abbdfbb1f91dc35280b1f90794 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 2 Jan 2014 21:09:39 -0800 Subject: Pretty: Added nestle. API change, minor version bump to 1.12.3. --- pandoc.cabal | 2 +- src/Text/Pandoc/Pretty.hs | 11 ++++++++++- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/pandoc.cabal b/pandoc.cabal index 976b71785..c794822bc 100644 --- a/pandoc.cabal +++ b/pandoc.cabal @@ -1,5 +1,5 @@ Name: pandoc -Version: 1.12.2.1 +Version: 1.12.3 Cabal-Version: >= 1.10 Build-Type: Custom License: GPL diff --git a/src/Text/Pandoc/Pretty.hs b/src/Text/Pandoc/Pretty.hs index faf2a6797..033511832 100644 --- a/src/Text/Pandoc/Pretty.hs +++ b/src/Text/Pandoc/Pretty.hs @@ -60,6 +60,7 @@ module Text.Pandoc.Pretty ( , hsep , vcat , vsep + , nestle , chomp , inside , braces @@ -72,7 +73,7 @@ module Text.Pandoc.Pretty ( ) where -import Data.Sequence (Seq, fromList, (<|), singleton, mapWithIndex) +import Data.Sequence (Seq, fromList, (<|), singleton, mapWithIndex, viewl, ViewL(..)) import Data.Foldable (toList) import Data.List (intercalate) import Data.Monoid @@ -186,6 +187,14 @@ vcat = foldr ($$) empty vsep :: [Doc] -> Doc vsep = foldr ($+$) empty +-- | Removes leading blank lines from a 'Doc'. +nestle :: Doc -> Doc +nestle (Doc d) = Doc $ go d + where go x = case viewl x of + (BlankLine :< rest) -> go rest + (NewLine :< rest) -> go rest + _ -> x + -- | Chomps trailing blank space off of a 'Doc'. chomp :: Doc -> Doc chomp d = Doc (fromList dl') -- cgit v1.2.3 From e7e76dbdd8b07b1fd80bda1599d58b7fbc7cf4bd Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 2 Jan 2014 21:10:14 -0800 Subject: RST writer: Ensure no blank line after def in definition list. Closes #992. --- src/Text/Pandoc/Writers/RST.hs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/Text') 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 -- cgit v1.2.3 From 2dd6d892fa310135cece1d8e6d6bac7b538f4e76 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Mon, 6 Jan 2014 10:17:31 -0800 Subject: HTML writer: Omit footnotes from TOC entries. Otherwise we get doubled footnotes when headers have notes! --- src/Text/Pandoc/Writers/HTML.hs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/HTML.hs b/src/Text/Pandoc/Writers/HTML.hs index c8a7e802b..805bb57f1 100644 --- a/src/Text/Pandoc/Writers/HTML.hs +++ b/src/Text/Pandoc/Writers/HTML.hs @@ -749,7 +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 -- cgit v1.2.3 From ba6a26b25871a3912556f1a17330d2363b3f3db5 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Mon, 6 Jan 2014 12:12:21 -0800 Subject: EPUB writer: Avoid duplicate notes when headings contain notes. This arose because the headings are copied into the metadata "title" field, and the note gets rendered twice. We strip the note now before putting the heading in "title". --- src/Text/Pandoc/Writers/EPUB.hs | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index 4daa9609e..d2dd7da2e 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -397,6 +397,10 @@ writeEPUB opts doc@(Pandoc meta _) = do let chapters = evalState (toChapters blocks'') [] + let removeNote :: Inline -> Inline + removeNote (Note _) = Str "" + removeNote x = x + let chapToEntry :: Int -> Chapter -> Entry chapToEntry num (Chapter mbnum bs) = mkEntry (showChapter num) $ renderHtml @@ -404,7 +408,9 @@ writeEPUB opts doc@(Pandoc meta _) = do 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 -- cgit v1.2.3 From d97b1fd14cd2b0f96fcf3bcc10cfc2540fbbe624 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Mon, 6 Jan 2014 21:51:11 -0800 Subject: EPUB writer: Strip out footnotes from toc entries. --- src/Text/Pandoc/Writers/EPUB.hs | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/EPUB.hs b/src/Text/Pandoc/Writers/EPUB.hs index d2dd7da2e..a48300939 100644 --- a/src/Text/Pandoc/Writers/EPUB.hs +++ b/src/Text/Pandoc/Writers/EPUB.hs @@ -124,7 +124,12 @@ opfName n = QName n Nothing (Just "opf") plainify :: [Inline] -> String plainify t = - trimr $ writePlain def{ writerStandalone = False } $ Pandoc nullMeta [Plain 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 @@ -397,10 +402,6 @@ writeEPUB opts doc@(Pandoc meta _) = do let chapters = evalState (toChapters blocks'') [] - let removeNote :: Inline -> Inline - removeNote (Note _) = Str "" - removeNote x = x - let chapToEntry :: Int -> Chapter -> Entry chapToEntry num (Chapter mbnum bs) = mkEntry (showChapter num) $ renderHtml -- cgit v1.2.3 From 002c8ce14ca15bfbc800e9628cf09babf3d96acd Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Tue, 7 Jan 2014 09:01:32 -0800 Subject: Fixed small regression in docx writer. --- src/Text/Pandoc/Writers/Docx.hs | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Writers/Docx.hs b/src/Text/Pandoc/Writers/Docx.hs index 25739f7c8..2a834c2da 100644 --- a/src/Text/Pandoc/Writers/Docx.hs +++ b/src/Text/Pandoc/Writers/Docx.hs @@ -30,7 +30,7 @@ Conversion of 'Pandoc' documents to docx. -} module Text.Pandoc.Writers.Docx ( writeDocx ) where import Data.Maybe (fromMaybe) -import Data.List ( intercalate, isPrefixOf ) +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 @@ -265,6 +265,7 @@ writeDocx opts doc@(Pandoc meta _) = do 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 @@ -815,6 +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" -- cgit v1.2.3 From 2c7bf41d26dfe293f32830d92d5bfd13f2cf2b89 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Tue, 7 Jan 2014 10:32:47 -0800 Subject: Added wmf and emf mime types. --- src/Text/Pandoc/MIME.hs | 2 ++ 1 file changed, 2 insertions(+) (limited to 'src/Text') diff --git a/src/Text/Pandoc/MIME.hs b/src/Text/Pandoc/MIME.hs index f41aa98bb..44989ee94 100644 --- a/src/Text/Pandoc/MIME.hs +++ b/src/Text/Pandoc/MIME.hs @@ -147,6 +147,7 @@ mimeTypesList = -- List borrowed from happstack-server. ,("dxr","application/x-director") ,("emb","chemical/x-embl-dl-nucleotide") ,("embl","chemical/x-embl-dl-nucleotide") + ,("emf","image/x-emf") ,("eml","message/rfc822") ,("ent","chemical/x-ncbi-asn1-ascii") ,("eot","application/vnd.ms-fontobject") @@ -465,6 +466,7 @@ mimeTypesList = -- List borrowed from happstack-server. ,("wm","video/x-ms-wm") ,("wma","audio/x-ms-wma") ,("wmd","application/x-ms-wmd") + ,("wmf","image/x-wmf") ,("wml","text/vnd.wap.wml") ,("wmlc","application/vnd.wap.wmlc") ,("wmls","text/vnd.wap.wmlscript") -- cgit v1.2.3 From d9eff99f27bb6426534651c9ff85b2b4354e6d16 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Tue, 7 Jan 2014 23:39:49 -0800 Subject: Markdown reader: Allow hard line breaks in table cells. The \-newline form must be used; the two-space+newline form won't work, since in a table cell nearly every line ends with two spaces. --- src/Text/Pandoc/Readers/Markdown.hs | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 3a5d29b4e..88d0bf439 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1126,12 +1126,12 @@ multilineTableHeader headless = try $ do then liftM (map (:[]) . tail . splitStringByIndices (init indices)) $ lookAhead anyLine else return $ transpose $ map - (\ln -> tail $ splitStringByIndices (init indices) ln) + (tail . splitStringByIndices (init indices)) rawContent let aligns = zipWith alignType rawHeadsList lengths let rawHeads = if headless then replicate (length dashes) "" - else map unwords rawHeadsList + else map (unlines . map trim) rawHeadsList heads <- fmap sequence $ mapM (parseFromString (mconcat <$> many plain)) $ map trim rawHeads @@ -1188,7 +1188,7 @@ gridTableHeader headless = try $ do -- RST does not have a notion of alignments let rawHeads = if headless then replicate (length dashes) "" - else map unwords $ transpose + else map (unlines . map trim) $ transpose $ map (gridTableSplitLine indices) rawContent heads <- fmap sequence $ mapM (parseFromString parseBlocks . trim) rawHeads return (heads, aligns, indices) -- cgit v1.2.3 From aada7b495bf4af9912603b3b7649dd0d63f9b5fc Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Wed, 8 Jan 2014 12:04:08 -0800 Subject: fetchItem: Handle image URLs beginning with '//'. --- src/Text/Pandoc/Shared.hs | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Shared.hs b/src/Text/Pandoc/Shared.hs index 3446f4343..714402e42 100644 --- a/src/Text/Pandoc/Shared.hs +++ b/src/Text/Pandoc/Shared.hs @@ -91,7 +91,8 @@ import Data.Char ( toLower, isLower, isUpper, isAlpha, isLetter, isDigit, isSpace ) import Data.List ( find, isPrefixOf, intercalate ) import qualified Data.Map as M -import Network.URI ( escapeURIString, isURI, unEscapeString ) +import Network.URI ( escapeURIString, isURI, nonStrictRelativeTo, + unEscapeString, parseURIReference ) import System.Directory import Text.Pandoc.MIME (getMimeType) import System.FilePath ( (</>), takeExtension, dropExtension ) @@ -623,9 +624,13 @@ fetchItem :: Maybe String -> String -> IO (Either E.SomeException (BS.ByteString, Maybe String)) fetchItem sourceURL s | isURI s = openURL s - | otherwise = case sourceURL of - Just u -> openURL (u ++ "/" ++ s) - Nothing -> E.try readLocalFile + | otherwise = + case sourceURL >>= parseURIReference of + Just u -> case parseURIReference s of + Just s' -> openURL $ show $ + s' `nonStrictRelativeTo` u + Nothing -> openURL $ show u ++ "/" ++ s + Nothing -> E.try readLocalFile where readLocalFile = do let mime = case takeExtension s of ".gz" -> getMimeType $ dropExtension s -- cgit v1.2.3 From 3bf8012bf6e6965a68de76d5bb46782086393da7 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Wed, 8 Jan 2014 19:33:14 -0800 Subject: Text.Pandoc.ImageSize: Parse EXIF format JPGs. Note: For now we just assign them all 72 dpi. It wasn't clear to me how to extract the resolution information. At least the aspect ratio will be right, and 72 dpi is the most common setting. Closes #976. --- src/Text/Pandoc/ImageSize.hs | 34 ++++++++++++++++++++++++++++------ 1 file changed, 28 insertions(+), 6 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/ImageSize.hs b/src/Text/Pandoc/ImageSize.hs index 9b0850efb..e2a8b8283 100644 --- a/src/Text/Pandoc/ImageSize.hs +++ b/src/Text/Pandoc/ImageSize.hs @@ -53,7 +53,8 @@ imageType :: ByteString -> Maybe ImageType imageType img = case B.take 4 img of "\x89\x50\x4e\x47" -> return Png "\x47\x49\x46\x38" -> return Gif - "\xff\xd8\xff\xe0" -> return Jpeg + "\xff\xd8\xff\xe0" -> return Jpeg -- JFIF + "\xff\xd8\xff\xe1" -> return Jpeg -- Exif "%PDF" -> return Pdf "%!PS" | (B.take 4 $ B.drop 1 $ B.dropWhile (/=' ') img) == "EPSF" @@ -139,8 +140,14 @@ gifSize img = do jpegSize :: ByteString -> Maybe ImageSize jpegSize img = do let (hdr, rest) = B.splitAt 4 img - guard $ hdr == "\xff\xd8\xff\xe0" guard $ B.length rest >= 14 + case hdr of + "\xff\xd8\xff\xe0" -> jfifSize rest + "\xff\xd8\xff\xe1" -> exifSize rest + _ -> mzero + +jfifSize :: ByteString -> Maybe ImageSize +jfifSize rest = do let [dpiDensity,dpix1,dpix2,dpiy1,dpiy2] = map fromIntegral $ unpack $ B.take 5 $ B.drop 9 $ rest let factor = case dpiDensity of @@ -149,11 +156,11 @@ jpegSize img = do _ -> const 72 let dpix = factor (shift dpix1 8 + dpix2) let dpiy = factor (shift dpiy1 8 + dpiy2) - (w,h) <- findJpegSize rest + (w,h) <- findJfifSize rest return $ ImageSize { pxX = w, pxY = h, dpiX = dpix, dpiY = dpiy } -findJpegSize :: ByteString -> Maybe (Integer,Integer) -findJpegSize bs = do +findJfifSize :: ByteString -> Maybe (Integer,Integer) +findJfifSize bs = do let bs' = B.dropWhile (=='\xff') $ B.dropWhile (/='\xff') bs case B.uncons bs' of Just (c,bs'') | c >= '\xc0' && c <= '\xc3' -> do @@ -165,8 +172,23 @@ findJpegSize bs = do [c1,c2] -> do let len = shift c1 8 + c2 -- skip variables - findJpegSize $ B.drop len bs'' + findJfifSize $ B.drop len bs'' _ -> fail "JPEG parse error" Nothing -> fail "Did not find length record" +exifSize :: ByteString -> Maybe ImageSize +exifSize rest = do + let bs' = B.takeWhile (/='\xff') $ B.drop 8 rest -- exif data + let (_,bs'') = B.breakSubstring "\xa0\x02" bs' -- width + let rawWidth = B.take 2 $ B.drop 10 bs'' + let (_,bs''') = B.breakSubstring "\xa0\x03" bs' -- height + let rawHeight = B.take 2 $ B.drop 10 bs''' + let tonum bs = case map fromIntegral $ unpack bs of + [x,y] -> Just $ shift x 8 + y + _ -> Nothing + case (tonum rawWidth, tonum rawHeight) of + (Just w, Just h) -> + return $ ImageSize { pxX = w, pxY = h, dpiX = 72, dpiY = 72 } + _ -> fail "Could not determine exif image size" + -- some day, figure out how to parse dpi from exif -- cgit v1.2.3 From 5c8c380a7997156964a5402974f6f464233aab9b Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 9 Jan 2014 11:16:17 -0800 Subject: Better exif parsing, including image resolution. This introduces a dependency on binary >= 0.6, but we depend on binary >= 0.5 via zip-archive anyway. Closes #976. --- pandoc.cabal | 3 +- src/Text/Pandoc/ImageSize.hs | 225 ++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 212 insertions(+), 16 deletions(-) (limited to 'src/Text') diff --git a/pandoc.cabal b/pandoc.cabal index a1a4c9b40..94d382c4f 100644 --- a/pandoc.cabal +++ b/pandoc.cabal @@ -232,7 +232,8 @@ Library attoparsec >= 0.10 && < 0.11, yaml >= 0.8.3 && < 0.9, vector >= 0.10 && < 0.11, - hslua >= 0.3 && < 0.4 + hslua >= 0.3 && < 0.4, + binary >= 0.6 && < 0.8 Build-Tools: alex, happy if flag(http-conduit) Build-Depends: http-conduit >= 1.9 && < 2.1, diff --git a/src/Text/Pandoc/ImageSize.hs b/src/Text/Pandoc/ImageSize.hs index e2a8b8283..467205220 100644 --- a/src/Text/Pandoc/ImageSize.hs +++ b/src/Text/Pandoc/ImageSize.hs @@ -32,9 +32,14 @@ module Text.Pandoc.ImageSize ( ImageType(..), imageType, imageSize, sizeInPixels, sizeInPoints ) where import Data.ByteString (ByteString, unpack) import qualified Data.ByteString.Char8 as B +import qualified Data.ByteString.Lazy as BL +import Control.Applicative import Control.Monad import Data.Bits +import Data.Binary +import Data.Binary.Get import Text.Pandoc.Shared (safeRead) +import qualified Data.Map as M -- quick and dirty functions to get image sizes -- algorithms borrowed from wwwis.pl @@ -143,7 +148,7 @@ jpegSize img = do guard $ B.length rest >= 14 case hdr of "\xff\xd8\xff\xe0" -> jfifSize rest - "\xff\xd8\xff\xe1" -> exifSize rest + "\xff\xd8\xff\xe1" -> exifSize $ B.takeWhile (/= '\xff') rest _ -> mzero jfifSize :: ByteString -> Maybe ImageSize @@ -177,18 +182,208 @@ findJfifSize bs = do Nothing -> fail "Did not find length record" exifSize :: ByteString -> Maybe ImageSize -exifSize rest = do - let bs' = B.takeWhile (/='\xff') $ B.drop 8 rest -- exif data - let (_,bs'') = B.breakSubstring "\xa0\x02" bs' -- width - let rawWidth = B.take 2 $ B.drop 10 bs'' - let (_,bs''') = B.breakSubstring "\xa0\x03" bs' -- height - let rawHeight = B.take 2 $ B.drop 10 bs''' - let tonum bs = case map fromIntegral $ unpack bs of - [x,y] -> Just $ shift x 8 + y - _ -> Nothing - case (tonum rawWidth, tonum rawHeight) of - (Just w, Just h) -> - return $ ImageSize { pxX = w, pxY = h, dpiX = 72, dpiY = 72 } - _ -> fail "Could not determine exif image size" - -- some day, figure out how to parse dpi from exif +exifSize = -- runGet ((Just <$> exifHeader) `mplus` return Nothing) . + runGet (Just <$> exifHeader) . + BL.fromChunks . (:[]) +exifHeader :: Get ImageSize +exifHeader = do + _app1DataSize <- getWord16be + exifHdr <- getWord32be + unless (exifHdr == 0x45786966) $ fail "Did not find exif header" + zeros <- getWord16be + unless (zeros == 0) $ fail "Expected zeros after exif header" + -- beginning of tiff header -- we read whole thing to use + -- in getting data from offsets: + tiffHeader <- lookAhead getRemainingLazyByteString + byteAlign <- getWord16be + let bigEndian = byteAlign == 0x4d4d + let (getWord16, getWord32, getWord64) = + if bigEndian + then (getWord16be, getWord32be, getWord64be) + else (getWord16le, getWord32le, getWord64le) + let getRational = do + num <- getWord32 + den <- getWord32 + return $ fromIntegral num / fromIntegral den + tagmark <- getWord16 + unless (tagmark == 0x002a) $ fail "Failed alignment sanity check" + ifdOffset <- getWord32 + skip (fromIntegral ifdOffset - 8) -- skip to IDF + numentries <- getWord16 + let ifdEntry = do + tag <- getWord16 >>= \t -> + maybe (fail $ "Unknown tag type " ++ show t) return + (M.lookup t tagTypeTable) + dataFormat <- getWord16 + numComponents <- getWord32 + (fmt, bytesPerComponent) <- + case dataFormat of + 1 -> return (UnsignedByte . runGet getWord8, 1) + 2 -> return (AsciiString, 1) + 3 -> return (UnsignedShort . runGet getWord16, 2) + 4 -> return (UnsignedLong . runGet getWord32, 4) + 5 -> return (UnsignedRational . runGet getRational, 8) + 6 -> return (SignedByte . runGet getWord8, 1) + 7 -> return (Undefined . runGet getWord8, 1) + 8 -> return (SignedShort . runGet getWord16, 2) + 9 -> return (SignedLong . runGet getWord32, 4) + 10 -> return (SignedRational . runGet getRational, 8) + 11 -> return (SingleFloat . runGet getWord32 {- TODO -}, 4) + 12 -> return (DoubleFloat . runGet getWord64 {- TODO -}, 8) + _ -> fail $ "Unknown data format " ++ show dataFormat + let totalBytes = fromIntegral $ numComponents * bytesPerComponent + payload <- if totalBytes <= 4 -- data is right here + then (fmt . BL.fromChunks . (:[])) <$> + (getByteString totalBytes <* + skip (4 - totalBytes)) + else do -- get data from offset + offs <- getWord32 + return $ fmt $ BL.take (fromIntegral totalBytes) $ + BL.drop (fromIntegral offs) tiffHeader + return (tag, payload) + entries <- sequence $ replicate (fromIntegral numentries) ifdEntry + subentries <- case lookup ExifOffset entries of + Just (UnsignedLong offset) -> do + pos <- bytesRead + skip (fromIntegral offset - (fromIntegral pos - 8)) + numsubentries <- getWord16 + sequence $ + replicate (fromIntegral numsubentries) ifdEntry + _ -> return [] + let allentries = entries ++ subentries + (width, height) <- case (lookup ExifImageWidth allentries, + lookup ExifImageHeight allentries) of + (Just (UnsignedLong w), Just (UnsignedLong h)) -> + return (fromIntegral w, fromIntegral h) + _ -> fail "Could not determine image width, height" + let resfactor = case lookup ResolutionUnit allentries of + Just (UnsignedShort 1) -> (100 / 254) + _ -> 1 + let xres = maybe 72 (\(UnsignedRational x) -> floor $ x * resfactor) + $ lookup XResolution allentries + let yres = maybe 72 (\(UnsignedRational x) -> floor $ x * resfactor) + $ lookup YResolution allentries + return $ ImageSize{ + pxX = width + , pxY = height + , dpiX = xres + , dpiY = yres } + +data DataFormat = UnsignedByte Word8 + | AsciiString BL.ByteString + | UnsignedShort Word16 + | UnsignedLong Word32 + | UnsignedRational Rational + | SignedByte Word8 + | Undefined Word8 + | SignedShort Word16 + | SignedLong Word32 + | SignedRational Rational + | SingleFloat Word32 + | DoubleFloat Word64 + deriving (Show) + +data TagType = ImageDescription + | Make + | Model + | Orientation + | XResolution + | YResolution + | ResolutionUnit + | Software + | DateTime + | WhitePoint + | PrimaryChromaticities + | YCbCrCoefficients + | YCbCrPositioning + | ReferenceBlackWhite + | Copyright + | ExifOffset + | ExposureTime + | FNumber + | ExposureProgram + | ISOSpeedRatings + | ExifVersion + | DateTimeOriginal + | DateTimeDigitized + | ComponentConfiguration + | CompressedBitsPerPixel + | ShutterSpeedValue + | ApertureValue + | BrightnessValue + | ExposureBiasValue + | MaxApertureValue + | SubjectDistance + | MeteringMode + | LightSource + | Flash + | FocalLength + | MakerNote + | UserComment + | FlashPixVersion + | ColorSpace + | ExifImageWidth + | ExifImageHeight + | RelatedSoundFile + | ExifInteroperabilityOffset + | FocalPlaneXResolution + | FocalPlaneYResolution + | FocalPlaneResolutionUnit + | SensingMethod + | FileSource + | SceneType + deriving (Show, Eq, Ord) + +tagTypeTable :: M.Map Word16 TagType +tagTypeTable = M.fromList + [ (0x010e, ImageDescription) + , (0x010f, Make) + , (0x0110, Model) + , (0x0112, Orientation) + , (0x011a, XResolution) + , (0x011b, YResolution) + , (0x0128, ResolutionUnit) + , (0x0131, Software) + , (0x0132, DateTime) + , (0x013e, WhitePoint) + , (0x013f, PrimaryChromaticities) + , (0x0211, YCbCrCoefficients) + , (0x0213, YCbCrPositioning) + , (0x0214, ReferenceBlackWhite) + , (0x8298, Copyright) + , (0x8769, ExifOffset) + , (0x829a, ExposureTime) + , (0x829d, FNumber) + , (0x8822, ExposureProgram) + , (0x8827, ISOSpeedRatings) + , (0x9000, ExifVersion) + , (0x9003, DateTimeOriginal) + , (0x9004, DateTimeDigitized) + , (0x9101, ComponentConfiguration) + , (0x9102, CompressedBitsPerPixel) + , (0x9201, ShutterSpeedValue) + , (0x9202, ApertureValue) + , (0x9203, BrightnessValue) + , (0x9204, ExposureBiasValue) + , (0x9205, MaxApertureValue) + , (0x9206, SubjectDistance) + , (0x9207, MeteringMode) + , (0x9208, LightSource) + , (0x9209, Flash) + , (0x920a, FocalLength) + , (0x927c, MakerNote) + , (0x9286, UserComment) + , (0xa000, FlashPixVersion) + , (0xa001, ColorSpace) + , (0xa002, ExifImageWidth) + , (0xa003, ExifImageHeight) + , (0xa004, RelatedSoundFile) + , (0xa005, ExifInteroperabilityOffset) + , (0xa20e, FocalPlaneXResolution) + , (0xa20f, FocalPlaneYResolution) + , (0xa210, FocalPlaneResolutionUnit) + , (0xa217, SensingMethod) + , (0xa300, FileSource) + , (0xa301, SceneType) + ] -- cgit v1.2.3 From b9b1546ed238ca8fd4d65a6e02fa7ecbf9a4be65 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 9 Jan 2014 11:25:24 -0800 Subject: Markdown parser: be more permissive about citation keys. Keys may now start with an underscore as well as a letter. Underscores do not count as internal punctuation, but are treated like alphanumerics, so "key:_2008" will work, as it did not before. (This change was necessary to use keys generated by zotero.) Closes #1111, closes #1011. --- src/Text/Pandoc/Readers/Markdown.hs | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 88d0bf439..3feafd362 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1811,9 +1811,10 @@ citeKey = try $ do guard $ lastStrPos /= Just pos suppress_author <- option False (char '-' >> return True) char '@' - first <- letter - let internal p = try $ p >>~ lookAhead (letter <|> digit) - rest <- many $ letter <|> digit <|> internal (oneOf ":.#$%&-_+?<>~/") + first <- letter <|> char '_' + let regchar = satisfy (\c -> isAlphaNum c || c == '_') + let internal p = try $ p >>~ lookAhead regchar + rest <- many $ regchar <|> internal (oneOf ":.#$%&-+?<>~/") let key = first:rest return (suppress_author, key) -- cgit v1.2.3 From b4b16d57865b1d951be79f04f8ad41fb70544077 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Thu, 9 Jan 2014 22:50:51 -0800 Subject: Minor improvement to exif parser. --- src/Text/Pandoc/ImageSize.hs | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/ImageSize.hs b/src/Text/Pandoc/ImageSize.hs index 467205220..d1aacff1c 100644 --- a/src/Text/Pandoc/ImageSize.hs +++ b/src/Text/Pandoc/ImageSize.hs @@ -234,8 +234,8 @@ exifHeader = do _ -> fail $ "Unknown data format " ++ show dataFormat let totalBytes = fromIntegral $ numComponents * bytesPerComponent payload <- if totalBytes <= 4 -- data is right here - then (fmt . BL.fromChunks . (:[])) <$> - (getByteString totalBytes <* + then fmt <$> + (getLazyByteString (fromIntegral totalBytes) <* skip (4 - totalBytes)) else do -- get data from offset offs <- getWord32 -- cgit v1.2.3 From a1abb3eeea2321654a8450725ff6c0d1a18ee0c7 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Tue, 14 Jan 2014 10:12:33 -0800 Subject: Allow binary 0.5. Version bump to 1.12.3.1. --- changelog | 4 ++++ pandoc.cabal | 4 ++-- src/Text/Pandoc/ImageSize.hs | 9 ++++++--- 3 files changed, 12 insertions(+), 5 deletions(-) (limited to 'src/Text') diff --git a/changelog b/changelog index 4cbc60388..fde6873ec 100644 --- a/changelog +++ b/changelog @@ -1,3 +1,7 @@ +pandoc (1.12.3.1) + + * Relaxed version constraint on binary, allowing the use of binary 0.5. + pandoc (1.12.3) * The `--bibliography` option now sets the `biblio-files` variable. diff --git a/pandoc.cabal b/pandoc.cabal index c8ff9738a..0199996be 100644 --- a/pandoc.cabal +++ b/pandoc.cabal @@ -1,5 +1,5 @@ Name: pandoc -Version: 1.12.3 +Version: 1.12.3.1 Cabal-Version: >= 1.10 Build-Type: Custom License: GPL @@ -233,7 +233,7 @@ Library yaml >= 0.8.3 && < 0.9, vector >= 0.10 && < 0.11, hslua >= 0.3 && < 0.4, - binary >= 0.6 && < 0.8 + binary >= 0.5 && < 0.8 Build-Tools: alex, happy if flag(http-conduit) Build-Depends: http-conduit >= 1.9 && < 2.1, diff --git a/src/Text/Pandoc/ImageSize.hs b/src/Text/Pandoc/ImageSize.hs index d1aacff1c..14575244d 100644 --- a/src/Text/Pandoc/ImageSize.hs +++ b/src/Text/Pandoc/ImageSize.hs @@ -182,9 +182,12 @@ findJfifSize bs = do Nothing -> fail "Did not find length record" exifSize :: ByteString -> Maybe ImageSize -exifSize = -- runGet ((Just <$> exifHeader) `mplus` return Nothing) . - runGet (Just <$> exifHeader) . - BL.fromChunks . (:[]) +exifSize = runGet (Just <$> exifHeader) . BL.fromChunks . (:[]) +-- NOTE: It would be nicer to do +-- runGet ((Just <$> exifHeader) <|> return Nothing) +-- which would prevent pandoc from raising an error when an exif header can't +-- be parsed. But we only get an Alternative instance for Get in binary 0.6, +-- and binary 0.5 ships with ghc 7.6. exifHeader :: Get ImageSize exifHeader = do -- cgit v1.2.3 From 6c59f060a769c7f312137d5c1a49cf878ea5b2a2 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Mon, 20 Jan 2014 11:09:44 -0800 Subject: HTML reader: Fixed bug reading inline math with `$$`. See #225. --- src/Text/Pandoc/Readers/HTML.hs | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index 506fe7770..d1e4d0024 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -476,8 +476,8 @@ pBlank = try $ do pTagContents :: Parser [Char] ParserState Inline pTagContents = - Math InlineMath `fmap` mathInline - <|> Math DisplayMath `fmap` mathDisplay + Math DisplayMath `fmap` mathDisplay + <|> Math InlineMath `fmap` mathInline <|> pStr <|> pSpace <|> smartPunctuation pTagContents -- cgit v1.2.3 From 9f3b2f6f5d06a4cf3142ffc74c8de4c1cc2bd928 Mon Sep 17 00:00:00 2001 From: John MacFarlane <jgm@berkeley.edu> Date: Wed, 22 Jan 2014 22:07:13 -0800 Subject: Fixed mediawiki ordered list parsing. Closes #1122. --- src/Text/Pandoc/Readers/MediaWiki.hs | 25 +++++++++++++++---------- tests/mediawiki-reader.wiki | 4 ++-- 2 files changed, 17 insertions(+), 12 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/MediaWiki.hs b/src/Text/Pandoc/Readers/MediaWiki.hs index 8d8ea0199..794890eb6 100644 --- a/src/Text/Pandoc/Readers/MediaWiki.hs +++ b/src/Text/Pandoc/Readers/MediaWiki.hs @@ -149,9 +149,16 @@ inlinesInTags tag = try $ do blocksInTags :: String -> MWParser Blocks blocksInTags tag = try $ do (_,raw) <- htmlTag (~== TagOpen tag []) + let closer = if tag == "li" + then htmlTag (~== TagClose "li") + <|> lookAhead ( + htmlTag (~== TagOpen "li" []) + <|> htmlTag (~== TagClose "ol") + <|> htmlTag (~== TagClose "ul")) + else htmlTag (~== TagClose tag) if '/' `elem` raw -- self-closing tag then return mempty - else mconcat <$> manyTill block (htmlTag (~== TagClose tag)) + else mconcat <$> manyTill block closer charsInTags :: String -> MWParser [Char] charsInTags tag = try $ do @@ -381,15 +388,13 @@ bulletList = B.bulletList <$> orderedList :: MWParser Blocks orderedList = (B.orderedList <$> many1 (listItem '#')) - <|> (B.orderedList <$> (htmlTag (~== TagOpen "ul" []) *> spaces *> - many (listItem '#' <|> li) <* - optional (htmlTag (~== TagClose "ul")))) - <|> do (tag,_) <- htmlTag (~== TagOpen "ol" []) - spaces - items <- many (listItem '#' <|> li) - optional (htmlTag (~== TagClose "ol")) - let start = fromMaybe 1 $ safeRead $ fromAttrib "start" tag - return $ B.orderedListWith (start, DefaultStyle, DefaultDelim) items + <|> try + (do (tag,_) <- htmlTag (~== TagOpen "ol" []) + spaces + items <- many (listItem '#' <|> li) + optional (htmlTag (~== TagClose "ol")) + let start = fromMaybe 1 $ safeRead $ fromAttrib "start" tag + return $ B.orderedListWith (start, DefaultStyle, DefaultDelim) items) definitionList :: MWParser Blocks definitionList = B.definitionList <$> many1 defListItem diff --git a/tests/mediawiki-reader.wiki b/tests/mediawiki-reader.wiki index 26f4ef164..c0c22bec6 100644 --- a/tests/mediawiki-reader.wiki +++ b/tests/mediawiki-reader.wiki @@ -232,11 +232,11 @@ ends the list. <li>list item A2</li> </ol> -<ul> +<ol> #abc #def #ghi -</ul> +</ol> <ol start="9"> <li>Amsterdam</li> -- cgit v1.2.3 From a333d9788e0f510f681ae1b5f0f246434ee15d62 Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Fri, 24 Jan 2014 16:00:02 -0800 Subject: ImageSize: Avoid use of lookAhead, which is not in binary >= 0.6. Closes #1124. --- src/Text/Pandoc/ImageSize.hs | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/ImageSize.hs b/src/Text/Pandoc/ImageSize.hs index 14575244d..3c9623b3c 100644 --- a/src/Text/Pandoc/ImageSize.hs +++ b/src/Text/Pandoc/ImageSize.hs @@ -182,15 +182,16 @@ findJfifSize bs = do Nothing -> fail "Did not find length record" exifSize :: ByteString -> Maybe ImageSize -exifSize = runGet (Just <$> exifHeader) . BL.fromChunks . (:[]) +exifSize bs = runGet (Just <$> exifHeader bl) bl + where bl = BL.fromChunks [bs] -- NOTE: It would be nicer to do -- runGet ((Just <$> exifHeader) <|> return Nothing) -- which would prevent pandoc from raising an error when an exif header can't -- be parsed. But we only get an Alternative instance for Get in binary 0.6, -- and binary 0.5 ships with ghc 7.6. -exifHeader :: Get ImageSize -exifHeader = do +exifHeader :: BL.ByteString -> Get ImageSize +exifHeader hdr = do _app1DataSize <- getWord16be exifHdr <- getWord32be unless (exifHdr == 0x45786966) $ fail "Did not find exif header" @@ -198,7 +199,7 @@ exifHeader = do unless (zeros == 0) $ fail "Expected zeros after exif header" -- beginning of tiff header -- we read whole thing to use -- in getting data from offsets: - tiffHeader <- lookAhead getRemainingLazyByteString + let tiffHeader = BL.drop 8 hdr byteAlign <- getWord16be let bigEndian = byteAlign == 0x4d4d let (getWord16, getWord32, getWord64) = -- cgit v1.2.3 From 3127ab2b5ede10224f9f86722125573d34d7d3dd Mon Sep 17 00:00:00 2001 From: John MacFarlane <fiddlosopher@gmail.com> Date: Tue, 4 Feb 2014 10:05:52 -0800 Subject: Slight code reorganization in endline. --- src/Text/Pandoc/Readers/Markdown.hs | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Readers/Markdown.hs b/src/Text/Pandoc/Readers/Markdown.hs index 3feafd362..8a41cef49 100644 --- a/src/Text/Pandoc/Readers/Markdown.hs +++ b/src/Text/Pandoc/Readers/Markdown.hs @@ -1540,16 +1540,14 @@ endline :: MarkdownParser (F Inlines) endline = try $ do newline notFollowedBy blankline + -- parse potential list-starts differently if in a list: + st <- getState + when (stateParserContext st == ListItemState) $ notFollowedBy listStart guardDisabled Ext_lists_without_preceding_blankline <|> notFollowedBy listStart guardEnabled Ext_blank_before_blockquote <|> notFollowedBy emailBlockQuoteStart guardEnabled Ext_blank_before_header <|> notFollowedBy (char '#') -- atx header guardEnabled Ext_backtick_code_blocks >> notFollowedBy (() <$ (lookAhead (char '`') >> codeBlockFenced)) - -- parse potential list-starts differently if in a list: - st <- getState - when (stateParserContext st == ListItemState) $ do - notFollowedBy' bulletListStart - notFollowedBy' anyOrderedListStart (guardEnabled Ext_hard_line_breaks >> return (return B.linebreak)) <|> (guardEnabled Ext_ignore_line_breaks >> return mempty) <|> (return $ return B.space) -- cgit v1.2.3 From 286781f8014cd40ad741e52b254904ffa7dc2855 Mon Sep 17 00:00:00 2001 From: Merijn Verstraaten <merijn@inconsistent.nl> Date: Thu, 6 Feb 2014 23:10:59 +0100 Subject: Removed RenderState datatype context. Reasoning: - It's not Haskell2010 - It breaks some tools - Doesn't actually do anything - RenderState doesn't even have a Monoid instance --- src/Text/Pandoc/Pretty.hs | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'src/Text') diff --git a/src/Text/Pandoc/Pretty.hs b/src/Text/Pandoc/Pretty.hs index 033511832..5331587ce 100644 --- a/src/Text/Pandoc/Pretty.hs +++ b/src/Text/Pandoc/Pretty.hs @@ -81,8 +81,7 @@ import Data.String import Control.Monad.State import Data.Char (isSpace) -data Monoid a => - RenderState a = RenderState{ +data RenderState a = RenderState{ output :: [a] -- ^ In reverse order , prefix :: String , usePrefix :: Bool -- cgit v1.2.3