diff options
| author | John MacFarlane <jgm@berkeley.edu> | 2010-12-03 22:30:08 -0800 |
|---|---|---|
| committer | John MacFarlane <jgm@berkeley.edu> | 2010-12-03 23:10:52 -0800 |
| commit | 36d4aa4a0998f35d4b7a9deab92a1f661d50dd9a (patch) | |
| tree | 0d462ebc970e31b1a26ee2e1eb4cc70355419506 | |
| parent | 55e43c49916e55c4fec9a6783590b9574a183bf9 (diff) | |
| download | pandoc-36d4aa4a0998f35d4b7a9deab92a1f661d50dd9a.tar.gz | |
Textile reader: modified str to handle acronyms, hyphens.
* A single hyphen between two word characters is no longer a
potential strikeout-starter.
* Acronym explanations are dropped.
| -rw-r--r-- | src/Text/Pandoc/Readers/Textile.hs | 19 |
1 files changed, 16 insertions, 3 deletions
diff --git a/src/Text/Pandoc/Readers/Textile.hs b/src/Text/Pandoc/Readers/Textile.hs index 276f188c5..52eebd07f 100644 --- a/src/Text/Pandoc/Readers/Textile.hs +++ b/src/Text/Pandoc/Readers/Textile.hs @@ -70,7 +70,8 @@ import Text.Pandoc.Readers.HTML ( htmlTag, htmlEndTag, -- find code blocks rawHtmlBlock, rawHtmlInline ) import Text.Pandoc.Readers.Markdown (smartPunctuation) import Text.ParserCombinators.Parsec -import Data.Char ( digitToInt ) +import Data.Char ( digitToInt, isLetter ) +import Control.Monad ( guard ) -- | Parse a Textile text and return a Pandoc document. readTextile :: ParserState -- ^ Parser state, including options for parser @@ -312,7 +313,19 @@ inlineParsers = [ autoLink -- | Any string str :: GenParser Char ParserState Inline -str = many1 (noneOf (specialChars ++ "\t\n ")) >>= return . Str +str = do + xs <- many1 (noneOf (specialChars ++ "\t\n ")) + optional $ charsInBalanced '(' ')' -- drop acronym explanation + -- e.g. PBS(Public Broadcasting Service) + -- parse a following hyphen if followed by a letter + -- (this prevents unwanted interpretation as starting a strikeout section) + result <- option xs $ try $ do + guard $ not . null $ xs + char '-' + next <- lookAhead letter + guard $ isLetter (last xs) || isLetter next + return $ xs ++ "-" + return $ Str result -- | Textile allows HTML span infos, we discard them htmlSpan :: GenParser Char ParserState Inline @@ -384,4 +397,4 @@ simpleInline :: GenParser Char ParserState t -- ^ surrounding parser -> GenParser Char ParserState Inline -- ^ content parser (to be used repeatedly) simpleInline border construct = surrounded border (inlineWithAttribute) >>= return . construct . normalizeSpaces - where inlineWithAttribute = (try $ optional attributes) >> inline
\ No newline at end of file + where inlineWithAttribute = (try $ optional attributes) >> inline |