JATS writer: use <break/> for LineBreak...

in the limited contexts that accept it. Closes #8344.
2026-08-31 02:04:53 +08:00 · 2022-10-01 17:46:13 -07:00
parent 89e7e9b636
commit dd07d8002f
4 changed files with 109 additions and 7 deletions
@@ -135,8 +135,11 @@ docToJATS opts (Pandoc meta blocks) = do
                            formatTime defaultTimeLocale "%F" day)
                      ]
          Just x -> x
+  title' <- inlinesToJATS opts $ map fixLineBreak
+               (lookupMetaInlines "title" meta)
  let context = defField "body" main
              $ defField "back" back
+              $ resetField "title" title'
              $ resetField "date" date
              $ defField "mathml" (case writerHTMLMathMethod opts of
                                        MathML -> True
@@ -249,6 +252,12 @@ codeAttr opts (ident,classes,kvs) = (lang, attr)
                    "platforms", "position", "specific-use"]]
       lang  = languageFor opts classes

+-- <break/> is only allowed as a direct child of <td> or <title> or
+-- <article-title>
+fixLineBreak :: Inline -> Inline
+fixLineBreak LineBreak = RawInline (Format "jats") "<break/>"
+fixLineBreak x = x
+
 -- | Convert a Pandoc block element to JATS.
 blockToJATS :: PandocMonad m => WriterOptions -> Block -> JATS m (Doc Text)
 blockToJATS _ Null = return empty
@@ -257,7 +266,7 @@ blockToJATS opts (Div (id',"section":_,kvs) (Header _lvl _ ils : xs)) = do
               | not (T.null id')]
  let otherAttrs = ["sec-type", "specific-use"]
  let attribs = idAttr ++ [(k,v) | (k,v) <- kvs, k `elem` otherAttrs]
-  title' <- inlinesToJATS opts ils
+  title' <- inlinesToJATS opts (map fixLineBreak ils)
  contents <- blocksToJATS opts xs
  return $ inTags True "sec" attribs $
      inTagsSimple "title" title' $$ contents
@@ -287,7 +296,7 @@ blockToJATS opts (Div (ident,_,kvs) bs) = do
                 "content-type", "orientation", "position"]]
  return $ inTags True "boxed-text" attr contents
 blockToJATS opts (Header _ _ title) = do
-  title' <- inlinesToJATS opts title
+  title' <- inlinesToJATS opts (map fixLineBreak title)
  return $ inTagsSimple "title" title'
 -- No Plain, everything needs to be in a block-level tag
 blockToJATS opts (Plain lst) = blockToJATS opts (Para lst)
@@ -234,8 +234,13 @@ tableCellToJats :: PandocMonad m
 tableCellToJats opts ctype colAlign (Cell attr align rowspan colspan item) = do
  blockToJats   <- asks jatsBlockWriter
  inlinesToJats <- asks jatsInlinesWriter
+  let fixBreak LineBreak = RawInline (Format "jats") "<break/>"
+      fixBreak x         = x
  let cellContents = \case
-        [Plain inlines] -> inlinesToJats opts inlines
+        [Plain inlines] -> inlinesToJats opts
+                             (map fixBreak inlines)
+                             -- Note: <break/> is allowed only as a direct
+                             -- child of <td>, so we don't use walk.
        blocks          -> blockToJats needsWrapInCell opts blocks
  let tag' = case ctype of
        BodyCell   -> "td"
@@ -0,0 +1,90 @@
+```
+% pandoc -t jats -s
+---
+title: |-
+  My\
+  document
+...
+
+# Section\
+with line break
+
+Paragraph\
+with line break
+
+--------
+A    B
+---  ---
+1\   3
+2    4
+
+*1\  5\
+2*   6
+
+     7
+--------
+
+^D
+<?xml version="1.0" encoding="utf-8" ?>
+<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.2 20190208//EN"
+                  "JATS-archivearticle1.dtd">
+<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" dtd-version="1.2" article-type="other">
+<front>
+<journal-meta>
+<journal-id></journal-id>
+<journal-title-group>
+</journal-title-group>
+<issn></issn>
+<publisher>
+<publisher-name></publisher-name>
+</publisher>
+</journal-meta>
+<article-meta>
+<title-group>
+<article-title>My<break/>document</article-title>
+</title-group>
+<permissions>
+</permissions>
+</article-meta>
+</front>
+<body>
+<sec id="section">
+  <title>Section<break/></title>
+  <p>with line break</p>
+  <p>Paragraph
+  with line break</p>
+  <table-wrap>
+    <table>
+      <colgroup>
+        <col width="7%" />
+        <col width="7%" />
+      </colgroup>
+      <thead>
+        <tr>
+          <th align="left">A</th>
+          <th align="left">B</th>
+        </tr>
+      </thead>
+      <tbody>
+        <tr>
+          <td align="left">1<break/>2</td>
+          <td align="left">3 4</td>
+        </tr>
+        <tr>
+          <td align="left"><italic>1
+          2</italic></td>
+          <td align="left">5<break/>6</td>
+        </tr>
+        <tr>
+          <td align="left"></td>
+          <td align="left">7</td>
+        </tr>
+      </tbody>
+    </table>
+  </table-wrap>
+</sec>
+</body>
+<back>
+</back>
+</article>
+```
@@ -13,10 +13,8 @@
      <tr>
        <th align="center">Name</th>
        <th align="center">Capital</th>
-        <th align="center">Population
-        (in 2018)</th>
-        <th align="center">Area
-        (in km<sup>2</sup>)</th>
+        <th align="center">Population<break/>(in 2018)</th>
+        <th align="center">Area<break/>(in km<sup>2</sup>)</th>
      </tr>
    </thead>
    <tfoot>