+ Improved text wrapping algorithm in markdown, docbook, and RST writers.

LineBreaks no longer cause ugly wrapping in Markdown output.
+ Replaced splitBySpace with the more general, polymorphic function
  splitBy (in Text/Pandoc/Shared).


git-svn-id: https://pandoc.googlecode.com/svn/trunk@411 788f1e2b-df1e-0410-8736-df70ead52e1b

debian/changelog

@@ -103,6 +103,8 @@ pandoc (0.3) unstable; urgency=low
     + Modified HTML reader to skip a newline following a <br> tag.
       Otherwise the newline will be treated as a space at the beginning
       of the next line.
+    + Fixed bug in text-wrapping routine in Markdown and RST writers.
+      Now LineBreaks no longer cause wrapping problems.
 
   * Made handling of code blocks more consistent.  Previously, some
     readers allowed trailing newlines, while others stripped them.
@@ -162,7 +164,8 @@ pandoc (0.3) unstable; urgency=low
       a '.exe' extension is added to each executable in EXECS.
 
   * Removed all wrappers except markdown2pdf and html2markdown.
-    Added new wrapper hsmarkdown, to be used as a drop-in replacement
+
+  * Added new wrapper hsmarkdown, to be used as a drop-in replacement
     for Markdown.pl.  hsmarkdown calls pandoc with the '--strict'
     option and disables other options.
 
@@ -185,6 +188,8 @@ pandoc (0.3) unstable; urgency=low
       not to be HTML-specific.
     + Removed 'BlockQuoteContext' from ParserContext, as it isn't
       used anywhere.
+    + Removed splitBySpace and replaced it with a general, polymorphic
+      splitBy function.
     + Refactored LaTeX reader for clarity (added isArg function).
     + Converted some CR's to LF's in src/ui/default/print.css.
     + Added license text to top of source files.

src/Text/Pandoc/Shared.hs

@@ -52,7 +52,7 @@ module Text.Pandoc.Shared (
                      -- * Pandoc block list processing
                      consolidateList,
                      isNoteBlock,
-                     splitBySpace,
+                     splitBy,
                      normalizeSpaces,
                      compactify,
                      generateReference,
@@ -274,10 +274,13 @@ removeTrailingSpace = reverse . removeLeadingSpace . reverse
 stripFirstAndLast str =
   drop 1 $ take ((length str) - 1) str
 
--- | Split list of inlines into groups separated by a space.
-splitBySpace :: [Inline] -> [[Inline]]
-splitBySpace lst = filter (\a -> (/= Space) (head a)) 
-                     (groupBy (\a b -> (/= Space) a && (/= Space) b) lst)
+-- | Split list into groups separated by sep.
+splitBy :: (Eq a) => a -> [a] -> [[a]]
+splitBy _ [] = []
+splitBy sep lst = 
+  let (first, rest) = break (== sep) lst
+      rest'         = dropWhile (== sep) rest in
+  first:(splitBy sep rest')
 
 -- | Normalize a list of inline elements: remove leading and trailing
 -- @Space@ elements, and collapse double @Space@s into singles.

src/Text/Pandoc/Writers/Docbook.hs

@@ -176,7 +176,7 @@ cdata str = text $ "<![CDATA[" ++ str ++ "]]>"
 
 -- | Take list of inline elements and return wrapped doc.
 wrap :: WriterOptions -> [Inline] -> Doc
-wrap options lst = fsep $ map (hcat . (map (inlineToDocbook options))) (splitBySpace lst)
+wrap options lst = fsep $ map (inlinesToDocbook options) (splitBy Space lst)
 
 -- | Escape a string for XML (with "smart" option if specified).
 stringToXML :: WriterOptions -> String -> String

src/Text/Pandoc/Writers/Markdown.hs

@@ -58,8 +58,13 @@ escapeLinkTitle = gsub "\"" "\\\\\""
 
 -- | Take list of inline elements and return wrapped doc.
 wrappedMarkdown :: [Inline] -> Doc
-wrappedMarkdown lst = fsep $ 
-                      map (fcat . (map inlineToMarkdown)) (splitBySpace lst) 
+wrappedMarkdown lst = 
+  let wrapSection sec = fsep $ map inlineListToMarkdown $ (splitBy Space sec) 
+      wrappedSecs     = map wrapSection $ splitBy LineBreak lst
+      wrappedSecs'    = foldr (\s rest -> if not (null rest)
+                                            then (s <> text "  "):rest
+                                            else s:rest) [] wrappedSecs  in
+  vcat wrappedSecs'
 
 -- | Insert Blank block between key and non-key
 formatKeys :: [Block] -> [Block]

src/Text/Pandoc/Writers/RST.hs

@@ -59,9 +59,10 @@ escapeString = backslashEscape "`\\|*_"
 -- and another containing references.
 wrappedRST :: [Inline] -> (Doc, Doc)
 wrappedRST lst = 
-  let words = splitBySpace lst in
-  ( fsep $ map (fcat . (map (fst . inlineToRST))) words, 
-    vcat (map (snd . inlineToRST) lst) )
+  let wrap_section sec = fsep $ map (fst . inlineListToRST) $ 
+                         (splitBy Space sec) in
+  ((vcat $ map wrap_section $ (splitBy LineBreak lst)),
+  vcat $ map (snd . inlineToRST) lst)
 
 -- | Remove reference keys, and make sure there are blanks before each list.
 reformatBlocks :: [Block] -> [Block]

tests/writer.rst

@@ -55,7 +55,8 @@ looked like a list item.
 
 Here's one with a bullet. \* criminey.
 
-There should be a hard line break here.
+There should be a hard line break
+here.
 
 --------------