JATS writer: fix XML output for nested figures

Subfigures are now wrapped inside a `<fig-group>` element. Furthermore,
figure content that isn't allowed as children of `<fig>` elements, such
as raw text, gets wrapped in `<p>` elements to ensure schema-conform
XML production.

Fixes: #11342

2 files changed, 63 insertions, 6 deletions

diff --git a/src/Text/Pandoc/Writers/JATS.hs b/src/Text/Pandoc/Writers/JATS.hs
index d7975bff3..2ac1497f3 100644
--- a/src/Text/Pandoc/Writers/JATS.hs
+++ b/src/Text/Pandoc/Writers/JATS.hs
@@ -439,12 +439,36 @@ blockToJATS opts (Figure (ident, _, kvs) (Caption _short longcapt) body) = do
   capt <- if null longcapt
           then pure empty
           else inTagsSimple "caption" <$> blocksToJATS opts longcapt
-  figbod <- blocksToJATS opts $ walk unsetAltIfDupl body
-  let figattr = [("id", escapeNCName ident) | not (T.null ident)] ++
-                [(k,v) | (k,v) <- kvs
-                       , k `elem` [ "fig-type", "orientation"
-                                  , "position", "specific-use"]]
-  return $ inTags True "fig" figattr $ capt $$ figbod
+
+  -- We handle the element specially if it's a figure with subfigures, i.e., if
+  -- all immediate children are figures themselves.
+  let hasSubfigures = all (\case { Figure{} -> True; _ -> False}) body
+      needsWrapping = if hasSubfigures
+                      then (const False)
+                      else \case
+                        -- Wrap all figure content elements, except for those
+                        -- allowed as direct subelements.
+                        BlockQuote{}    -> False
+                        CodeBlock{}     -> False
+                        Para{}          -> False
+                        Plain [Image{}] -> False
+                        Plain [Math{}]  -> False
+                        Table{}         -> False
+                        _               -> True
+
+  children <- wrappedBlocksToJATS needsWrapping opts $ walk unsetAltIfDupl body
+  let (tag, allowedAttributes) =
+        if hasSubfigures
+        then ( "fig-group"
+             , ["content-type", "orientation", "position", "specific-use"]
+             )
+        else ("fig"
+             , ["fig-type", "orientation", "position", "specific-use"]
+             )
+  let xmlattr =  [("id", escapeNCName ident) | not (T.null ident)] ++
+                 [(k,v) | (k,v) <- kvs
+                        , k `elem` allowedAttributes]
+  return $ inTags True tag xmlattr $ capt $$ children
 
 -- | Convert a list of inline elements to JATS.
 inlinesToJATS :: PandocMonad m => WriterOptions -> [Inline] -> JATS m (Doc Text)
diff --git a/test/command/11342.md b/test/command/11342.md
new file mode 100644
index 000000000..7e6df30ee
--- /dev/null
+++ b/test/command/11342.md
@@ -0,0 +1,33 @@
+Subfigures should be converted to *fig-group* elements.
+
+```
+% pandoc -f latex -t jats
+\begin{figure}[H]
+  \begin{subfigure}
+    \centering
+    \includegraphics[height=1.5in]{assets/A.png}
+    \caption{Graph for function *A*.}
+    \label{fig:A}
+  \end{subfigure}
+  \begin{subfigure}
+    \centering
+    \includegraphics[height=1.5in]{assets/B.png}
+    \caption{Graph for function *B*.}
+    \label{fig:B}
+  \end{subfigure}
+  \label{fig:graphs}
+  \caption{Function graphs.}
+\end{figure}
+^D
+<fig-group id="figU003Agraphs">
+  <caption><p>Function graphs.</p></caption>
+  <fig id="figU003AA">
+    <caption><p>Graph for function *A*.</p></caption>
+    <graphic mimetype="image" mime-subtype="png" xlink:href="assets/A.png" />
+  </fig>
+  <fig id="figU003AB">
+    <caption><p>Graph for function *B*.</p></caption>
+    <graphic mimetype="image" mime-subtype="png" xlink:href="assets/B.png" />
+  </fig>
+</fig-group>
+```
\ No newline at end of file