856aa8c · typeable/pandoc

Benchmark.hs+7−5

compare with master

unifiedsplit

@@ -3,6 +3,7 @@ import Text.Pandoc.Shared (readDataFile, normalize)

3import Criterion.Main

4import Data.List (isSuffixOf)

5import Text.JSON.Generic

6+import Data.Default

6

7

7readerBench :: Pandoc

8readerBench :: Pandoc

8 -> (String, ParserState -> String -> Pandoc)

9 -> (String, ParserState -> String -> Pandoc)

@@ -18,10 +19,11 @@ readerBench doc (name, reader) =

18 getLength (Pandoc (Meta a b c) d) =

19 getLength (Pandoc (Meta a b c) d) =

19 length a + length b + length c + length d

20 length a + length b + length c + length d

20 in bench (name ++ " reader") $ whnf (getLength .

21 in bench (name ++ " reader") $ whnf (getLength .

21− reader defaultParserState{ stateSmart = True

22+ reader def{ stateOptions = def{

22− , stateStandalone = True

23+ readerSmart = True

23− , stateLiterateHaskell =

24− "+lhs" `isSuffixOf` name }) inp

24+ , readerLiterateHaskell = "+lhs" `isSuffixOf` name

25+ }

26+ }) inp

25

27

26writerBench :: Pandoc

28writerBench :: Pandoc

27 -> (String, WriterOptions -> Pandoc -> String)

29 -> (String, WriterOptions -> Pandoc -> String)

@@ -38,7 +40,7 @@ normalizeBench doc = [ bench "normalize - with" $ nf (encodeJSON . normalize) do

38

40

39main = do

41main = do

40 inp <- readDataFile (Just ".") "README"

42 inp <- readDataFile (Just ".") "README"

41− let ps = defaultParserState{ stateSmart = True }

43+ let ps = defaultParserState{ stateOptions = def{ readerSmart = True }}

42 let doc = readMarkdown ps inp

44 let doc = readMarkdown ps inp

43 let readerBs = map (readerBench doc) readers

45 let readerBs = map (readerBench doc) readers

44 let writers' = [(n,w) | (n, PureStringWriter w) <- writers]

46 let writers' = [(n,w) | (n, PureStringWriter w) <- writers]

src/Tests/Readers/Markdown.hs+2−1

@@ -93,7 +93,8 @@ tests = [ testGroup "inline code"

93 =?> para (note (para "See [^1]"))

94 ]

95 , testGroup "lhs"

96− [ test (readMarkdown defaultParserState{stateLiterateHaskell = True})

96+ [ test (readMarkdown def{stateOptions =

97+ def{readerLiterateHaskell = True}})

97 "inverse bird tracks and html" $

98 "inverse bird tracks and html" $

98 "> a\n\n< b\n\n<div>\n"

99 "> a\n\n< b\n\n<div>\n"

99 =?> codeBlockWith ("",["sourceCode","literate","haskell"],[]) "a"

100 =?> codeBlockWith ("",["sourceCode","literate","haskell"],[]) "a"

src/Text/Pandoc.hs+9−3

@@ -170,16 +170,22 @@ readers = [("native" , \_ -> readNative)

170 ,("json" , \_ -> decodeJSON)

171 ,("markdown" , readMarkdown)

172 ,("markdown+lhs" , \st ->

173− readMarkdown st{ stateLiterateHaskell = True})

173+ readMarkdown st{ stateOptions =

174+ let oldopts = stateOptions st

175+ in oldopts{ readerLiterateHaskell = True} })

174 ,("rst" , readRST)

176 ,("rst" , readRST)

175 ,("rst+lhs" , \st ->

177 ,("rst+lhs" , \st ->

176− readRST st{ stateLiterateHaskell = True})

178+ readRST st{ stateOptions =

179+ let oldopts = stateOptions st

180+ in oldopts{ readerLiterateHaskell = True} })

177 ,("docbook" , readDocBook)

181 ,("docbook" , readDocBook)

178 ,("textile" , readTextile) -- TODO : textile+lhs

182 ,("textile" , readTextile) -- TODO : textile+lhs

179 ,("html" , readHtml)

183 ,("html" , readHtml)

180 ,("latex" , readLaTeX)

184 ,("latex" , readLaTeX)

181 ,("latex+lhs" , \st ->

185 ,("latex+lhs" , \st ->

182− readLaTeX st{ stateLiterateHaskell = True})

186+ readLaTeX st{ stateOptions =

187+ let oldopts = stateOptions st

188+ in oldopts{ readerLiterateHaskell = True} })

183 ]

189 ]

184

190

185data Writer = PureStringWriter (WriterOptions -> Pandoc -> String)

191data Writer = PureStringWriter (WriterOptions -> Pandoc -> String)

src/Text/Pandoc/Options.hs+16−14

@@ -51,24 +51,26 @@ data Extension = Footnotes

51 deriving (Show, Read, Enum, Eq, Ord, Bounded)

52

53data ReaderOptions = ReaderOptions{

54− readerExtensions :: Set Extension -- ^ Syntax extensions

54+ readerExtensions :: Set Extension -- ^ Syntax extensions

55− , readerSmart :: Bool -- ^ Smart punctuation

55+ , readerSmart :: Bool -- ^ Smart punctuation

56− , readerStrict :: Bool -- ^ FOR TRANSITION ONLY

56+ , readerStrict :: Bool -- ^ FOR TRANSITION ONLY

57− , readerParseRaw :: Bool -- ^ Parse raw HTML, LaTeX

57+ , readerParseRaw :: Bool -- ^ Parse raw HTML, LaTeX

58− , readerColumns :: Int -- ^ Number of columns in terminal

58+ , readerColumns :: Int -- ^ Number of columns in terminal

59− , readerTabStop :: Int -- ^ Tab stop

59+ , readerTabStop :: Int -- ^ Tab stop

60− , readerOldDashes :: Bool -- ^ Use pandoc <= 1.8.2.1 behavior

60+ , readerOldDashes :: Bool -- ^ Use pandoc <= 1.8.2.1 behavior

61 -- in parsing dashes; -- is em-dash;

62 -- - before numerial is en-dash

63+ , readerLiterateHaskell :: Bool -- ^ Interpret as literate Haskell

63} deriving (Show, Read)

64} deriving (Show, Read)

64

65

65instance Default ReaderOptions

66instance Default ReaderOptions

66 where def = ReaderOptions{

67 where def = ReaderOptions{

67− readerExtensions = Set.fromList [minBound..maxBound]

68+ readerExtensions = Set.fromList [minBound..maxBound]

68− , readerSmart = False

69+ , readerSmart = False

69− , readerStrict = False

70+ , readerStrict = False

70− , readerParseRaw = False

71+ , readerParseRaw = False

71− , readerColumns = 80

72+ , readerColumns = 80

72− , readerTabStop = 4

73+ , readerTabStop = 4

73− , readerOldDashes = False

74+ , readerOldDashes = False

75+ , readerLiterateHaskell = False

74 }

76 }

src/Text/Pandoc/Parsing.hs+1−3

@@ -395,7 +395,7 @@ failIfStrict = getOption readerStrict >>= guard . not

395

396-- | Fail unless we're in literate haskell mode.

397failUnlessLHS :: Parsec [tok] ParserState ()

398−failUnlessLHS = getState >>= guard . stateLiterateHaskell

398+failUnlessLHS = getOption readerLiterateHaskell >>= guard

399

400-- | Parses backslash, then applies character parser.

401escaped :: Parsec [Char] st Char -- ^ Parser for character to escape

@@ -698,7 +698,6 @@ data ParserState = ParserState

698 stateTitle :: [Inline], -- ^ Title of document

699 stateAuthors :: [[Inline]], -- ^ Authors of document

700 stateDate :: [Inline], -- ^ Date of document

701− stateLiterateHaskell :: Bool, -- ^ Treat input as literate haskell

702 stateHeaderTable :: [HeaderType], -- ^ Ordered list of header types used

701 stateHeaderTable :: [HeaderType], -- ^ Ordered list of header types used

703 stateIndentedCodeClasses :: [String], -- ^ Classes to use for indented code blocks

702 stateIndentedCodeClasses :: [String], -- ^ Classes to use for indented code blocks

704 stateNextExample :: Int, -- ^ Number of next example

703 stateNextExample :: Int, -- ^ Number of next example

@@ -726,7 +725,6 @@ defaultParserState =

726 stateTitle = [],

725 stateTitle = [],

727 stateAuthors = [],

726 stateAuthors = [],

728 stateDate = [],

727 stateDate = [],

729− stateLiterateHaskell = False,

730 stateHeaderTable = [],

728 stateHeaderTable = [],

731 stateIndentedCodeClasses = [],

729 stateIndentedCodeClasses = [],

732 stateNextExample = 1,

730 stateNextExample = 1,

src/pandoc.hs+4−3

@@ -936,9 +936,7 @@ main = do

936 then "."

937 else takeDirectory (head sources)

938

939− let startParserState = def{ stateLiterateHaskell = "+lhs" `isSuffixOf` readerName' ||

940− lhsExtension sources,

941− stateCitations = map CSL.refId refs,

939+ let startParserState = def{ stateCitations = map CSL.refId refs,

942 stateOptions = def{

940 stateOptions = def{

943 readerStrict = strict

941 readerStrict = strict

944 , readerSmart = smart || (texLigatures &&

942 , readerSmart = smart || (texLigatures &&

@@ -947,6 +945,9 @@ main = do

947 , readerColumns = columns

945 , readerColumns = columns

948 , readerTabStop = tabStop

946 , readerTabStop = tabStop

949 , readerOldDashes = oldDashes

947 , readerOldDashes = oldDashes

948+ , readerLiterateHaskell =

949+ "+lhs" `isSuffixOf` readerName' ||

950+ lhsExtension sources

950 },

951 },

951 stateIndentedCodeClasses = codeBlockClasses,

952 stateIndentedCodeClasses = codeBlockClasses,

952 stateApplyMacros = not laTeXOutput

953 stateApplyMacros = not laTeXOutput