e7eb21e · typeable/pandoc

git.iko.soy/typeable/pandoc

Lua module: add readers submodule

Plain text readers are exposed to lua scripts via the `pandoc.reader` submodule, which is further subdivided by format. Converting e.g. a markdown string into a pandoc document is possible from within lua: doc = pandoc.reader.markdown.read_doc("Hello, World!") A `read_block` convenience function is provided for all formats, although it will still parse the whole string but return only the first block as the result. Custom reader options are not supported yet, default options are used for all parsing operations.

commite7eb21ecca46daaf240e33584c55b9d5101eebc7 authorAlbert Krewinkel <albert@zeitkraut.de> dateSun 2 Apr 2017, 17:21 (10 years ago) parent9e78a9d

6 files changed+172−32

compare with master

unifiedsplit

pandoc.cabal+4−2 src/Text/Pandoc/Lua.hs+3−28 src/Text/Pandoc/Lua/PandocModule.hs+72−2 src/Text/Pandoc/Lua/StackInstances.hs+75 test/Tests/Lua.hs+6 test/lua/markdown-reader.lua+12

pandoc.cabal+4−2

@@ -247,9 +247,10 @@ Extra-Source-Files:

247247 test/odt/odt/*.odt

248248 test/odt/markdown/*.md

249249 test/odt/native/*.native

250− test/lua/strmacro.lua

251− test/lua/plain-to-para.lua

252250 test/lua/hello-world-doc.lua

251+ test/lua/markdown-reader.lua

252+ test/lua/plain-to-para.lua

253+ test/lua/strmacro.lua

253254Source-repository head

254255 type: git

255256 location: git://github.com/jgm/pandoc.git

@@ -458,6 +459,7 @@ Library

458459 Text.Pandoc.Readers.Org.Shared,

459460 Text.Pandoc.Lua.Compat,

460461 Text.Pandoc.Lua.PandocModule,

462+ Text.Pandoc.Lua.StackInstances,

461463 Text.Pandoc.CSS,

462464 Text.Pandoc.UUID,

463465 Text.Pandoc.Slides,

src/Text/Pandoc/Lua.hs+3−28

@@ -15,11 +15,7 @@ You should have received a copy of the GNU General Public License

1515along with this program; if not, write to the Free Software

1616Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

1717-}

18−{-# LANGUAGE FlexibleContexts #-}

19−{-# LANGUAGE LambdaCase #-}

2018{-# LANGUAGE OverloadedStrings #-}

21−{-# LANGUAGE ScopedTypeVariables #-}

22−{-# OPTIONS_GHC -fno-warn-orphans #-}

2319{- |

2420 Module : Text.Pandoc.Lua

@@ -34,24 +30,23 @@ module Text.Pandoc.Lua ( runLuaFilter ) where

3430

3531import Control.Monad ( (>=>), when )

3632import Control.Monad.Trans ( MonadIO(..) )

37−import Data.Aeson ( FromJSON(..), ToJSON(..), Result(..), Value, fromJSON )

3833import Data.HashMap.Lazy ( HashMap )

3934import Data.Text ( Text, pack, unpack )

4035import Data.Text.Encoding ( decodeUtf8 )

4136import Scripting.Lua ( LuaState, StackValue(..) )

42−import Scripting.Lua.Aeson ()

37+import Scripting.Lua.Aeson ( newstate )

4338import Text.Pandoc.Definition ( Block(..), Inline(..), Pandoc(..) )

4439import Text.Pandoc.Lua.PandocModule

40+import Text.Pandoc.Lua.StackInstances ()

4541import Text.Pandoc.Walk

4642

4743import qualified Data.HashMap.Lazy as HashMap

4844import qualified Scripting.Lua as Lua

49−import qualified Scripting.Lua as LuaAeson

5045

5146runLuaFilter :: (MonadIO m)

5247 => FilePath -> [String] -> Pandoc -> m Pandoc

5348runLuaFilter filterPath args pd = liftIO $ do

54− lua <- LuaAeson.newstate

49+ lua <- newstate

5550 Lua.openlibs lua

5651 Lua.newtable lua

5752 Lua.setglobal lua "PANDOC_FILTER_FUNCTIONS" -- hack, store functions here

@@ -204,23 +199,3 @@ isLuaFunction lua fnName = do

204199 res <- Lua.isfunction lua (-1)

205200 Lua.pop lua (-1)

206201 return res

207−

208−maybeFromJson :: (FromJSON a) => Maybe Value -> Maybe a

209−maybeFromJson mv = fromJSON <$> mv >>= \case

210− Success x -> Just x

211− _ -> Nothing

212−

213−instance StackValue Pandoc where

214− push lua = Lua.push lua . toJSON

215− peek lua i = maybeFromJson <$> peek lua i

216− valuetype _ = Lua.TTABLE

217−

218−instance StackValue Block where

219− push lua = Lua.push lua . toJSON

220− peek lua i = maybeFromJson <$> peek lua i

221− valuetype _ = Lua.TTABLE

222−

223−instance StackValue Inline where

224− push lua = Lua.push lua . toJSON

225− peek lua i = maybeFromJson <$> peek lua i

226− valuetype _ = Lua.TTABLE

src/Text/Pandoc/Lua/PandocModule.hs+72−2

@@ -28,11 +28,25 @@ Pandoc module for lua.

2828module Text.Pandoc.Lua.PandocModule ( pushPandocModule ) where

2929

3030import Data.ByteString.Char8 ( unpack )

31−import Scripting.Lua ( LuaState, call)

31+import Data.Default ( Default(..) )

32+import Scripting.Lua ( LuaState, call, newtable, push, pushhsfunction, rawset)

33+import Text.Pandoc.Class hiding ( readDataFile )

34+import Text.Pandoc.Definition ( Pandoc(..), Block(..) )

3235import Text.Pandoc.Lua.Compat ( loadstring )

36+import Text.Pandoc.Lua.StackInstances ()

37+import Text.Pandoc.Readers.DocBook ( readDocBook )

38+import Text.Pandoc.Readers.HTML ( readHtml )

39+import Text.Pandoc.Readers.LaTeX ( readLaTeX )

40+import Text.Pandoc.Readers.Native ( readNative )

41+import Text.Pandoc.Readers.Markdown ( readMarkdown )

42+import Text.Pandoc.Readers.MediaWiki ( readMediaWiki )

43+import Text.Pandoc.Readers.Org ( readOrg )

44+import Text.Pandoc.Readers.RST ( readRST )

45+import Text.Pandoc.Readers.Textile ( readTextile )

46+import Text.Pandoc.Readers.TWiki ( readTWiki )

47+import Text.Pandoc.Readers.Txt2Tags ( readTxt2Tags )

3348import Text.Pandoc.Shared ( readDataFile )

3449

35−

3650-- | Push the "pandoc" on the lua stack.

3751pushPandocModule :: LuaState -> IO ()

3852pushPandocModule lua = do

@@ -42,7 +56,63 @@ pushPandocModule lua = do

4256 then return ()

4357 else do

4458 call lua 0 1

59+ push lua "reader"

60+ pushReadersModule lua readers

61+ rawset lua (-3)

62+

63+readers :: [(String, String -> PandocIO Pandoc)]

64+readers =

65+ [ ("docbook", readDocBook def)

66+ , ("html", readHtml def)

67+ , ("latex", readLaTeX def)

68+ , ("native", readNative def)

69+ , ("markdown", readMarkdown def)

70+ , ("mediawiki", readMediaWiki def)

71+ , ("org", readOrg def)

72+ , ("rst", readRST def)

73+ , ("textile", readTextile def)

74+ , ("twiki", readTWiki def)

75+ , ("txt2tags", readTxt2Tags def)

76+ ]

4577

4678-- | Get the string representation of the pandoc module

4779pandocModuleScript :: IO String

4880pandocModuleScript = unpack <$> readDataFile Nothing "pandoc.lua"

81+

82+-- | Push a lua table containing readers of the given formats.

83+pushReadersModule :: LuaState

84+ -> [(String, String -> PandocIO Pandoc)]

85+ -> IO ()

86+pushReadersModule lua readerFns = do

87+ newtable lua

88+ mapM_ (uncurry $ addReaderTable) readerFns

89+ where

90+ addReaderTable :: String

91+ -> (String -> PandocIO Pandoc)

92+ -> IO ()

93+ addReaderTable formatName readerFn = do

94+ let readDoc :: String -> IO Pandoc

95+ readDoc s = do

96+ res <- runIO $ readerFn s

97+ case res of

98+ (Left x) -> error (show x)

99+ (Right x) -> return x

100+ let readBlock :: String -> IO Block

101+ readBlock s = do

102+ Pandoc _ blks <- readDoc s

103+ return $ case blks of

104+ x:_ -> x

105+ _ -> Null

106+ -- Push table containing all functions for this format

107+ push lua formatName

108+ newtable lua

109+ -- set document-reading function

110+ push lua "read_doc"

111+ pushhsfunction lua readDoc

112+ rawset lua (-3)

113+ -- set block-reading function

114+ push lua "read_block"

115+ pushhsfunction lua readBlock

116+ rawset lua (-3)

117+ -- store table in readers module

118+ rawset lua (-3)

src/Text/Pandoc/Lua/StackInstances.hsnew file+75

@@ -0,0 +1,75 @@

1+{-

3+ 2017 Albert Krewinkel <tarleb+pandoc@moltkeplatz.de>

5+This program is free software; you can redistribute it and/or modify

6+it under the terms of the GNU General Public License as published by

7+the Free Software Foundation; either version 2 of the License, or

8+(at your option) any later version.

10+This program is distributed in the hope that it will be useful,

11+but WITHOUT ANY WARRANTY; without even the implied warranty of

12+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

13+GNU General Public License for more details.

14+

15+You should have received a copy of the GNU General Public License

16+along with this program; if not, write to the Free Software

17+Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

18+-}

19+{-# LANGUAGE CPP #-}

20+{-# LANGUAGE FlexibleInstances #-}

21+{-# LANGUAGE LambdaCase #-}

22+#if !MIN_VERSION_base(4,8,0)

23+{-# LANGUAGE OverlappingInstances #-}

24+#endif

25+{-# OPTIONS_GHC -fno-warn-orphans #-}

26+{- |

27+ Module : Text.Pandoc.Lua.StackInstances

29+ License : GNU GPL, version 2 or above

30+

31+ Maintainer : Albert Krewinkel <tarleb+pandoc@moltkeplatz.de>

32+ Stability : alpha

33+

34+StackValue instances for pandoc types.

35+-}

36+module Text.Pandoc.Lua.StackInstances () where

37+

38+import Data.Aeson ( FromJSON(..), ToJSON(..), Result(..), Value, fromJSON )

39+import Scripting.Lua ( StackValue(..) )

40+import Scripting.Lua.Aeson ()

41+import Text.Pandoc.Definition ( Block(..), Inline(..), Pandoc(..) )

42+

43+import qualified Scripting.Lua as Lua

44+import qualified Text.Pandoc.UTF8 as UTF8

45+

46+maybeFromJson :: (FromJSON a) => Maybe Value -> Maybe a

47+maybeFromJson mv = fromJSON <$> mv >>= \case

48+ Success x -> Just x

49+ _ -> Nothing

50+

51+instance StackValue Pandoc where

52+ push lua = Lua.push lua . toJSON

53+ peek lua i = maybeFromJson <$> peek lua i

54+ valuetype _ = Lua.TTABLE

55+

56+instance StackValue Block where

57+ push lua = Lua.push lua . toJSON

58+ peek lua i = maybeFromJson <$> peek lua i

59+ valuetype _ = Lua.TTABLE

60+

61+instance StackValue Inline where

62+ push lua = Lua.push lua . toJSON

63+ peek lua i = maybeFromJson <$> peek lua i

64+ valuetype _ = Lua.TTABLE

65+

66+#if MIN_VERSION_base(4,8,0)

67+instance {-# OVERLAPS #-} StackValue [Char] where

68+#else

69+instance StackValue [Char] where

70+#endif

71+ push lua cs = Lua.push lua (UTF8.fromString cs)

72+ peek lua i = do

73+ res <- Lua.peek lua i

74+ return $ UTF8.toString `fmap` res

75+ valuetype _ = Lua.TSTRING

test/Tests/Lua.hs+6

@@ -26,6 +26,12 @@ tests =

2626 "hello-world-doc.lua"

2727 (doc . para $ str "Hey!" <> linebreak <> str "What's up?")

2828 (doc . para $ str "Hello," <> space <> str "World!")

29+

30+ , testCase "parse raw markdown blocks" $

31+ assertFilterConversion "raw markdown block is converted"

32+ "markdown-reader.lua"

33+ (doc $ rawBlock "markdown" "*charly* **delta**")

34+ (doc . para $ emph "charly" <> space <> strong "delta")

2935 ]

3036

3137assertFilterConversion :: String -> FilePath -> Pandoc -> Pandoc -> Assertion

test/lua/markdown-reader.luanew file+12

@@ -0,0 +1,12 @@

1+return {

2+ {

3+ RawBlock = function (blk)

4+ local format, content = unpack(blk.c)

5+ if format == "markdown" then

6+ return pandoc.reader.markdown.read_block(content)

7+ else

8+ return blk

9+ end

10+ end,

11+ }

12+}