Docx reader: read table styles as custom styles...

...when `styles` extension is enabled.  Closes #9603.

Also improve manual's coverage of custom styles.

6 files changed, 53 insertions, 18 deletions

diff --git a/MANUAL.txt b/MANUAL.txt
index 6ec998db9..7a7899319 100644
--- a/MANUAL.txt
+++ b/MANUAL.txt
@@ -3753,11 +3753,14 @@ output formats
 
 ### Extension: `styles` ### {#ext-styles}
 
-When converting from docx, read all docx styles as divs (for
-paragraph styles) and spans (for character styles) regardless
-of whether pandoc understands the meaning of these styles.
-This can be used with [docx custom styles](#custom-styles).
-Disabled by default.
+When converting from docx, add `custom-styles` attributes
+for all docx styles, regardless of whether pandoc understands
+the meanings of these styles. Because attributes cannot be
+added directly to paragraphs or text in the pandoc AST,
+paragraph styles will cause Divs to be created and character
+styles will cause Spans to be created to hold the attributes.
+(Table styles will be added to the Table elements directly.)
+This extension can be used with [docx custom styles](#custom-styles).
 
 input formats
 :  `docx`
@@ -7354,11 +7357,11 @@ However, if you need to apply your own styles to blocks, or match a preexisting
 set of styles, pandoc allows you to define custom styles for blocks and text
 using `div`s and `span`s, respectively.
 
-If you define a `div` or `span` with the attribute `custom-style`,
-pandoc will apply your specified style to the contained elements (with
-the exception of elements whose function depends on a style, like
-headings, code blocks, block quotes, or links). So, for example, using
-the `bracketed_spans` syntax,
+If you define a Div, Span, or Table with the attribute
+`custom-style`, pandoc will apply your specified style to the
+contained elements (with the exception of elements whose function
+depends on a style, like headings, code blocks, block quotes, or
+links). So, for example, using the `bracketed_spans` syntax,
 
     [Get out]{custom-style="Emphatically"}, he said.
 
@@ -7400,9 +7403,10 @@ interpreting the derivation of the input document's styles.
 
 By enabling the [`styles` extension](#ext-styles) in the docx reader
 (`-f docx+styles`), you can produce output that maintains the styles
-of the input document, using the `custom-style` class. Paragraph
-styles are interpreted as divs, while character styles are interpreted
-as spans.
+of the input document, using the `custom-style` class. A `custom-style`
+attribute will be added for each style. Divs will be created to
+hold the paragraph styles, and Spans to hold the character styles.
+Table styles will be applied directly to the Table.
 
 For example, using the `custom-style-reference.docx` file in the test
 directory, we have the following different outputs:
diff --git a/pandoc.cabal b/pandoc.cabal
index 08e863812..fdaa5caf2 100644
--- a/pandoc.cabal
+++ b/pandoc.cabal
@@ -219,6 +219,7 @@ extra-source-files:
                  test/command/9391.docx
                  test/command/9358.docx
                  test/command/9002.docx
+                 test/command/9603.docx
                  test/command/biblio.bib
                  test/command/averroes.bib
                  test/command/A.txt
diff --git a/src/Text/Pandoc/Readers/Docx.hs b/src/Text/Pandoc/Readers/Docx.hs
index 7abbc461d..efe104af7 100644
--- a/src/Text/Pandoc/Readers/Docx.hs
+++ b/src/Text/Pandoc/Readers/Docx.hs
@@ -807,9 +807,9 @@ bodyPartToBlocks (Captioned parstyle parparts bpart) = do
     [Para im@[Image{}]]
       -> pure $ singleton $ Figure nullAttr capt [Plain im]
     _ -> pure captContents
-bodyPartToBlocks (Tbl _ _ _ []) =
+bodyPartToBlocks (Tbl _ _ _ _ []) =
   return mempty
-bodyPartToBlocks (Tbl cap grid look parts) = do
+bodyPartToBlocks (Tbl mbsty cap grid look parts) = do
   let fullCaption = if T.null cap then mempty else plain (text cap)
   let shortCaption = if T.null cap then Nothing else Just (toList (text cap))
       cap' = caption shortCaption fullCaption
@@ -831,7 +831,11 @@ bodyPartToBlocks (Tbl cap grid look parts) = do
       totalWidth = sum grid
       widths = (\w -> ColWidth (fromInteger w / fromInteger totalWidth)) <$> grid
 
-  return $ table cap'
+  extStylesEnabled <- asks (isEnabled Ext_styles . docxOptions)
+  let attr = case mbsty of
+                Just sty | extStylesEnabled -> ("", [], [("custom-style", sty)])
+                _ -> nullAttr
+  return $ tableWith attr cap'
                  (zip alignments widths)
                  (TableHead nullAttr headerCells)
                  [TableBody nullAttr 0 [] bodyCells]
diff --git a/src/Text/Pandoc/Readers/Docx/Parse.hs b/src/Text/Pandoc/Readers/Docx/Parse.hs
index 20e510d8d..99ffcaf09 100644
--- a/src/Text/Pandoc/Readers/Docx/Parse.hs
+++ b/src/Text/Pandoc/Readers/Docx/Parse.hs
@@ -287,7 +287,7 @@ data BodyPart = Paragraph ParagraphStyle [ParPart]
               | Heading Int ParaStyleName ParagraphStyle T.Text T.Text (Maybe Level)
                  [ParPart]
               | ListItem ParagraphStyle T.Text T.Text (Maybe Level) [ParPart]
-              | Tbl T.Text TblGrid TblLook [Row]
+              | Tbl (Maybe T.Text) T.Text TblGrid TblLook [Row]
               | Captioned ParagraphStyle [ParPart] BodyPart
               | HRule
               deriving Show
@@ -855,6 +855,9 @@ elemToBodyPart ns element
         description = fromMaybe "" $ tblProperties
                        >>= findChildByName ns "w" "tblDescription"
                        >>= findAttrByName ns "w" "val"
+        mbstyle = tblProperties
+                       >>= findChildByName ns "w" "tblStyle"
+                       >>= findAttrByName ns "w" "val"
         grid' = case findChildByName ns "w" "tblGrid" element of
           Just g  -> elemToTblGrid ns g
           Nothing -> return []
@@ -867,7 +870,7 @@ elemToBodyPart ns element
     grid <- grid'
     tblLook <- tblLook'
     rows <- mapD (elemToRow ns) (elChildren element)
-    return $ Tbl (caption <> description) grid tblLook rows
+    return $ Tbl mbstyle (caption <> description) grid tblLook rows
 elemToBodyPart _ _ = throwError WrongElem
 
 lookupRelationship :: DocumentLocation -> RelId -> [Relationship] -> Maybe Target
diff --git a/test/command/9603.docx b/test/command/9603.docx
new file mode 100644
index 000000000..7d70a1249
--- /dev/null
+++ b/test/command/9603.docx
diff --git a/test/command/9603.md b/test/command/9603.md
new file mode 100644
index 000000000..fda2a777d
--- /dev/null
+++ b/test/command/9603.md
@@ -0,0 +1,23 @@
+```
+% pandoc command/9603.docx -t html -f docx+styles
+^D
+<p>A table with a contemporary style:</p>
+<table data-custom-style="TableContemporary">
+<colgroup>
+<col style="width: 50%" />
+<col style="width: 50%" />
+</colgroup>
+<thead>
+<tr>
+<th>Test</th>
+<th>Column</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>1</td>
+<td>2</td>
+</tr>
+</tbody>
+</table>
+```