ae8c0cd · typeable/pandoc

git.iko.soy/typeable/pandoc

on master help sign in

Docx reader: Parse instrText info in fldChar tags.

ae8c0cdJesse Rosenthal9 years ago

pandoc.cabal+1

compare with master

unifiedsplit

@@ -508,6 +508,7 @@ library

508 Text.Pandoc.Readers.Docx.Parse,

509 Text.Pandoc.Readers.Docx.Util,

510 Text.Pandoc.Readers.Docx.StyleMap,

511+ Text.Pandoc.Readers.Docx.Fields,

511 Text.Pandoc.Readers.Odt.Base,

512 Text.Pandoc.Readers.Odt.Base,

512 Text.Pandoc.Readers.Odt.Namespaces,

513 Text.Pandoc.Readers.Odt.Namespaces,

513 Text.Pandoc.Readers.Odt.StyleReader,

514 Text.Pandoc.Readers.Odt.StyleReader,

src/Text/Pandoc/Readers/Docx/Fields.hsnew file+89

@@ -0,0 +1,89 @@

1+{-

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

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

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

7+(at your option) any later version.

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

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

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

12+GNU General Public License for more details.

13+

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

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

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

17+-}

18+

19+{- |

20+ Module : Text.Pandoc.Readers.Docx.Fields

22+ License : GNU GPL, version 2 or above

23+

24+ Maintainer : Jesse Rosenthal <jrosenthal@jhu.edu>

25+ Stability : alpha

26+ Portability : portable

27+

28+For parsing Field definitions in instText tags, as described in

29+ECMA-376-1:2016, §17.16.5 -}

30+

31+module Text.Pandoc.Readers.Docx.Fields ( FieldInfo(..)

32+ , parseFieldInfo

33+ ) where

34+

35+import Text.Parsec

36+import Text.Parsec.String (Parser)

37+

38+type URL = String

39+

40+data FieldInfo = HyperlinkField URL

41+ | UnknownField

42+ deriving (Show)

43+

44+parseFieldInfo :: String -> Either ParseError FieldInfo

45+parseFieldInfo = parse fieldInfo ""

46+

47+fieldInfo :: Parser FieldInfo

48+fieldInfo =

49+ (try $ HyperlinkField <$> hyperlink)

50+ <|>

51+ return UnknownField

52+

53+escapedQuote :: Parser String

54+escapedQuote = string "\\\""

55+

56+inQuotes :: Parser String

57+inQuotes = do

58+ (try escapedQuote) <|> (anyChar >>= (\c -> return [c]))

59+

60+quotedString :: Parser String

61+quotedString = do

62+ char '"'

63+ concat <$> manyTill inQuotes (try (char '"'))

64+

65+unquotedString :: Parser String

66+unquotedString = manyTill anyChar (try (space))

67+

68+fieldArgument :: Parser String

69+fieldArgument = quotedString <|> unquotedString

70+

71+-- there are other switches, but this is the only one I've seen in the wild so far, so it's the first one I'll implement. See §17.16.5.25

72+hyperlinkSwitch :: Parser (String, String)

73+hyperlinkSwitch = do

74+ sw <- string "\\l"

75+ spaces

76+ farg <- fieldArgument

77+ return (sw, farg)

78+

79+hyperlink :: Parser URL

80+hyperlink = do

81+ many space

82+ string "HYPERLINK"

83+ spaces

84+ farg <- fieldArgument

85+ switches <- (spaces *> many hyperlinkSwitch)

86+ let url = case switches of

87+ ("\\l", s) : _ -> farg ++ ('#': s)

88+ _ -> farg

89+ return url

src/Text/Pandoc/Readers/Docx/Parse.hs+13−5

@@ -54,6 +54,7 @@ module Text.Pandoc.Readers.Docx.Parse ( Docx(..)

54 , TrackedChange(..)

55 , ChangeType(..)

56 , ChangeInfo(..)

57+ , FieldInfo(..)

57 , archiveToDocx

58 , archiveToDocx

58 , archiveToDocxWithWarnings

59 , archiveToDocxWithWarnings

59 ) where

60 ) where

@@ -70,6 +71,7 @@ import qualified Data.Map as M

70import Data.Maybe

71import Data.Maybe

71import System.FilePath

72import System.FilePath

72import Text.Pandoc.Readers.Docx.Util

73import Text.Pandoc.Readers.Docx.Util

74+import Text.Pandoc.Readers.Docx.Fields

73import Text.Pandoc.Shared (filteredFilesFromArchive, safeRead)

75import Text.Pandoc.Shared (filteredFilesFromArchive, safeRead)

74import qualified Text.Pandoc.UTF8 as UTF8

76import qualified Text.Pandoc.UTF8 as UTF8

75import Text.TeXMath (Exp)

77import Text.TeXMath (Exp)

@@ -286,10 +288,6 @@ data Run = Run RunStyle [RunElem]

286 | InlineChart -- placeholder

288 | InlineChart -- placeholder

287 deriving Show

289 deriving Show

288

290

289−data FieldInfo = HyperlinkField URL

290− | UnknownField

291− deriving (Show)

292−

293data RunElem = TextRun String | LnBrk | Tab | SoftHyphen | NoBreakHyphen

291data RunElem = TextRun String | LnBrk | Tab | SoftHyphen | NoBreakHyphen

294 deriving Show

292 deriving Show

295

293

@@ -802,8 +800,18 @@ elemToParPart ns element

802 return NullParPart

800 return NullParPart

803 FldCharContent info runs | fldCharType == "end" -> do

801 FldCharContent info runs | fldCharType == "end" -> do

804 modify $ \st -> st {stateFldCharState = FldCharClosed}

802 modify $ \st -> st {stateFldCharState = FldCharClosed}

805− return $ Field info runs

803+ return $ Field info $ reverse runs

806 _ -> throwError WrongElem

804 _ -> throwError WrongElem

805+elemToParPart ns element

806+ | isElem ns "w" "r" element

807+ , Just instrText <- findChildByName ns "w" "instrText" element = do

808+ fldCharState <- gets stateFldCharState

809+ case fldCharState of

810+ FldCharOpen -> do

811+ info <- eitherToD $ parseFieldInfo $ strContent instrText

812+ modify $ \st -> st{stateFldCharState = FldCharFieldInfo info}

813+ return NullParPart

814+ _ -> return NullParPart

807elemToParPart ns element

815elemToParPart ns element

808 | isElem ns "w" "r" element = do

816 | isElem ns "w" "r" element = do

809 run <- elemToRun ns element

817 run <- elemToRun ns element