95f2726 · typeable/pandoc

git.iko.soy/typeable/pandoc

Added readerAbbreviations to ParserState.

Markdown reader now consults this to determine what is an abbreviation. Eventually it will be possible to specify a custom list (see #256).

commit95f2726ee74a7770bf0eb5fe5629c493addbe298 authorJohn MacFarlane <jgm@berkeley.edu> dateSun 5 Mar 2017, 10:24 (10 years ago) parent7fc6919

2 files changed+27−25

compare with master

unifiedsplit

src/Text/Pandoc/Options.hs+11 src/Text/Pandoc/Readers/Markdown.hs+16−25

src/Text/Pandoc/Options.hs+11

@@ -47,6 +47,7 @@ module Text.Pandoc.Options ( module Text.Pandoc.Extensions

4747 ) where

4848import Data.Data (Data)

4949import Data.Default

50+import qualified Data.Set as Set

5051import Data.Typeable (Typeable)

5152import GHC.Generics (Generic)

5253import Text.Pandoc.Extensions

@@ -60,6 +61,7 @@ data ReaderOptions = ReaderOptions{

6061 , readerApplyMacros :: Bool -- ^ Apply macros to TeX math

6162 , readerIndentedCodeClasses :: [String] -- ^ Default classes for

6263 -- indented code blocks

64+ , readerAbbreviations :: Set.Set String -- ^ Strings to treat as abbreviations

6365 , readerDefaultImageExtension :: String -- ^ Default extension for images

6466 , readerTrackChanges :: TrackChanges

6567} deriving (Show, Read, Data, Typeable, Generic)

@@ -72,10 +74,19 @@ instance Default ReaderOptions

7274 , readerTabStop = 4

7375 , readerApplyMacros = True

7476 , readerIndentedCodeClasses = []

77+ , readerAbbreviations = defaultAbbrevs

7578 , readerDefaultImageExtension = ""

7679 , readerTrackChanges = AcceptChanges

7780 }

7881

82+defaultAbbrevs :: Set.Set String

83+defaultAbbrevs = Set.fromList

84+ [ "Mr.", "Mrs.", "Ms.", "Capt.", "Dr.", "Prof.",

85+ "Gen.", "Gov.", "e.g.", "i.e.", "Sgt.", "St.",

86+ "vol.", "vs.", "Sen.", "Rep.", "Pres.", "Hon.",

87+ "Rev.", "Ph.D.", "M.D.", "M.A.", "p.", "pp.",

88+ "ch.", "sec.", "cf.", "cp."]

89+

7990--

8091-- Writer options

8192--

src/Text/Pandoc/Readers/Markdown.hs+16−25

@@ -42,6 +42,7 @@ import Data.Maybe

4242import Data.Monoid ((<>))

4343import Data.Ord (comparing)

4444import Data.Scientific (base10Exponent, coefficient)

45+import qualified Data.Set as Set

4546import Data.Text (Text)

4647import qualified Data.Text as T

4748import qualified Data.Vector as V

@@ -1688,32 +1689,22 @@ nonEndline = satisfy (/='\n')

16881689

16891690str :: PandocMonad m => MarkdownParser m (F Inlines)

16901691str = do

1691− result <- many1 alphaNum

1692+ result <- many1 (alphaNum <|> try (char '.' <* notFollowedBy (char '.')))

16921693 updateLastStrPos

1693− let spacesToNbr = map (\c -> if c == ' ' then '\160' else c)

1694− isSmart <- extensionEnabled Ext_smart <$> getOption readerExtensions

1695− if isSmart

1696− then case likelyAbbrev result of

1697− [] -> return $ return $ B.str result

1698− xs -> choice (map (\x ->

1699− try (string x >> oneOf " \n" >>

1700− lookAhead alphaNum >>

1701− return (return $ B.str

1702− $ result ++ spacesToNbr x ++ "\160"))) xs)

1703− <|> (return $ return $ B.str result)

1704− else return $ return $ B.str result

1705−

1706−-- | if the string matches the beginning of an abbreviation (before

1707−-- the first period, return strings that would finish the abbreviation.

1708−likelyAbbrev :: String -> [String]

1709−likelyAbbrev x =

1710− let abbrevs = [ "Mr.", "Mrs.", "Ms.", "Capt.", "Dr.", "Prof.",

1711− "Gen.", "Gov.", "e.g.", "i.e.", "Sgt.", "St.",

1712− "vol.", "vs.", "Sen.", "Rep.", "Pres.", "Hon.",

1713− "Rev.", "Ph.D.", "M.D.", "M.A.", "p.", "pp.",

1714− "ch.", "sec.", "cf.", "cp."]

1715− abbrPairs = map (break (=='.')) abbrevs

1716− in map snd $ filter (\(y,_) -> y == x) abbrPairs

1694+ (do guardEnabled Ext_smart

1695+ abbrevs <- getOption readerAbbreviations

1696+ if not (null result) && last result == '.' && result `Set.member` abbrevs

1697+ then try (do ils <- whitespace <|> endline

1698+ lookAhead alphaNum

1699+ return $ do

1700+ ils' <- ils

1701+ if ils' == B.space

1702+ then return (B.str result <> B.str "\160")

1703+ else -- linebreak or softbreak

1704+ return (ils' <> B.str result <> B.str "\160"))

1705+ <|> return (return (B.str result))

1706+ else return (return (B.str result)))

1707+ <|> return (return (B.str result))

17171708

17181709-- an endline character that can be treated as a space, not a structural break

17191710endline :: PandocMonad m => MarkdownParser m (F Inlines)