[odt] Infer table's caption from the paragraph (#3224)

ODT's reader always put empty captions for the parsed
tables. This commit
1) checks paragraphs that follow the table definition
2) treats specially a paragraph with a style named 'Table'
3) does some postprocessing of the paragraphs that combines
 tables followed immediately by captions

The ODT writer used 'TableCaption' style name for the caption
paragraph. This commit follows the open office approach which
allows for appending captions to table but uses a built-in style
named 'Table' instead of 'TableCaption'. Any users of odt format
(both writer and reader) are therefore required to change the
style's name to 'Table', if necessary.

data/odt/styles.xml

@@ -133,7 +133,7 @@ xmlns:css3t="http://www.w3.org/TR/css3-text/" office:version="1.2">
       style:font-size-complex="12pt"
       style:font-style-complex="italic" />
     </style:style>
-    <style:style style:name="TableCaption" style:family="paragraph"
+    <style:style style:name="Table" style:family="paragraph"
     style:parent-style-name="Caption" style:class="extra">
     </style:style>
     <style:style style:name="FigureCaption" style:family="paragraph"

src/Text/Pandoc/Readers/Odt/ContentReader.hs

@@ -312,7 +312,7 @@ withNewStyle a = proc x -> do
   where
     isCodeStyle :: StyleName -> Bool
     isCodeStyle "Source_Text" = True
-    isCodeStyle _              = False
+    isCodeStyle _             = False
 
     inlineCode :: Inlines -> Inlines
     inlineCode = code . intercalate "" . map stringify . toList
@@ -415,12 +415,18 @@ constructPara reader = proc blocks -> do
   fStyle <- readStyleByName -< blocks
   case fStyle of
     Left   _    -> reader -< blocks
+    Right (styleName, _) | isTableCaptionStyle styleName -> do
+      blocks' <- reader   -< blocks
+      arr tableCaptionP  -< blocks'
     Right (_, style) -> do
       let modifier = getParaModifier style
-      blocks' <- reader -< blocks
-      arr modifier -<< blocks'
-
-
+      blocks' <- reader   -<  blocks
+      arr modifier        -<< blocks'
+  where
+    isTableCaptionStyle :: StyleName -> Bool
+    isTableCaptionStyle "Table" = True
+    isTableCaptionStyle _       = False
+    tableCaptionP b = divWith ("", ["caption"], []) b
 
 type ListConstructor = [Blocks] -> Blocks
 
@@ -904,6 +910,15 @@ read_text = matchChildContent' [ read_header
                                ]
             >>^ doc
 
+post_process :: Pandoc -> Pandoc
+post_process (Pandoc m blocks) =
+  Pandoc m (post_process' blocks)
+
+post_process' :: [Block] -> [Block]
+post_process' ((Table _ a w h r) : (Div ("", ["caption"], _) [Para inlines] ) : xs) =
+  (Table inlines a w h r) : ( post_process' xs )
+post_process' bs = bs
+
 read_body :: OdtReader _x (Pandoc, MediaBag)
 read_body = executeIn NsOffice "body"
           $ executeIn NsOffice "text"
@@ -911,4 +926,4 @@ read_body = executeIn NsOffice "body"
           $ proc inlines -> do
              txt   <- read_text     -< inlines
              state <- getExtraState -< ()
-             returnA                -< (txt, getMediaBag state)
+             returnA                -< (post_process txt, getMediaBag state)

src/Text/Pandoc/Writers/OpenDocument.hs

@@ -336,7 +336,7 @@ blockToOpenDocument o bs
         mapM_ addParaStyle . newPara $ paraHStyles ++ paraStyles
         captionDoc <- if null c
                       then return empty
-                      else withParagraphStyle o "TableCaption" [Para c]
+                      else withParagraphStyle o "Table" [Para c]
         th <- if all null h
                  then return empty
                  else colHeadsToOpenDocument o name (map fst paraHStyles) h

tests/Tests/Readers/Odt.hs

@@ -157,6 +157,7 @@ namesOfTestsComparingToNative   = [ "blockquote"
                                   , "referenceToListItem"
                                   , "referenceToText"
                                   , "simpleTable"
+                                  , "simpleTableWithCaption"
 --                                , "table"
                                   , "unicode"
                                   , "unorderedList"

tests/odt/native/simpleTableWithCaption.native

@@ -0,0 +1 @@
+[Table [Str "Table",Space,Str "1:",Space,Str "Some",Space,Str "caption",Space,Str "for",Space,Str "a",Space,Str "table"] [AlignDefault,AlignDefault] [0.0,0.0] [[],[]] [[[Plain [Str "Content"]],[Plain [Str "More",Space,Str "content"]]]],Para []]

tests/tables.opendocument

@@ -63,8 +63,7 @@
     </table:table-cell>
   </table:table-row>
 </table:table>
-<text:p text:style-name="TableCaption">Demonstration of simple table
-syntax.</text:p>
+<text:p text:style-name="Table">Demonstration of simple table syntax.</text:p>
 <text:p text:style-name="First_20_paragraph">Simple table without
 caption:</text:p>
 <table:table table:name="Table2" table:style-name="Table2">
@@ -197,8 +196,7 @@ spaces:</text:p>
     </table:table-cell>
   </table:table-row>
 </table:table>
-<text:p text:style-name="TableCaption">Demonstration of simple table
-syntax.</text:p>
+<text:p text:style-name="Table">Demonstration of simple table syntax.</text:p>
 <text:p text:style-name="First_20_paragraph">Multiline table with
 caption:</text:p>
 <table:table table:name="Table4" table:style-name="Table4">
@@ -253,8 +251,8 @@ caption:</text:p>
     </table:table-cell>
   </table:table-row>
 </table:table>
-<text:p text:style-name="TableCaption">Here's the caption. It may span
-multiple lines.</text:p>
+<text:p text:style-name="Table">Here's the caption. It may span multiple
+lines.</text:p>
 <text:p text:style-name="First_20_paragraph">Multiline table without
 caption:</text:p>
 <table:table table:name="Table5" table:style-name="Table5">