JATS writer: escape disallows chars in identifiers

XML identifiers must start with an underscore or letter, and can contain
only a limited set of punctuation characters. Any IDs not adhering to
these rules are rewritten by writing the offending characters as Uxxxx,
where `xxxx` is the character's hex code.

2 files changed, 5 insertions, 4 deletions

diff --git a/src/Text/Pandoc/Writers/JATS/References.hs b/src/Text/Pandoc/Writers/JATS/References.hs
index 903144128..5b19fd034 100644
--- a/src/Text/Pandoc/Writers/JATS/References.hs
+++ b/src/Text/Pandoc/Writers/JATS/References.hs
@@ -29,7 +29,7 @@ import Text.Pandoc.Builder (Inlines)
 import Text.Pandoc.Options (WriterOptions)
 import Text.Pandoc.Shared (tshow)
 import Text.Pandoc.Writers.JATS.Types
-import Text.Pandoc.XML (escapeStringForXML, inTags)
+import Text.Pandoc.XML (escapeNCName, escapeStringForXML, inTags)
 import qualified Data.Text as T
 
 referencesToJATS :: PandocMonad m
@@ -46,7 +46,8 @@ referenceToJATS :: PandocMonad m
 referenceToJATS _opts ref = do
   let refType = referenceType ref
   let pubType = [("publication-type", refType) | not (T.null refType)]
-  let wrap = inTags True "ref" [("id", "ref-" <> unItemId (referenceId ref))]
+  let ident = escapeNCName $ "ref-" <> unItemId (referenceId ref)
+  let wrap = inTags True "ref" [("id", ident)]
            . inTags True "element-citation" pubType
   return . wrap . vcat $
     [ authors
diff --git a/src/Text/Pandoc/Writers/JATS/Table.hs b/src/Text/Pandoc/Writers/JATS/Table.hs
index 465480f59..2e34900d2 100644
--- a/src/Text/Pandoc/Writers/JATS/Table.hs
+++ b/src/Text/Pandoc/Writers/JATS/Table.hs
@@ -24,7 +24,7 @@ import Text.Pandoc.Definition
 import Text.Pandoc.Options (WriterOptions)
 import Text.Pandoc.Shared (tshow)
 import Text.Pandoc.Writers.JATS.Types
-import Text.Pandoc.XML (inTags, inTagsIndented, selfClosingTag)
+import Text.Pandoc.XML (escapeNCName, inTags, inTagsIndented, selfClosingTag)
 import qualified Data.Text as T
 import qualified Text.Pandoc.Writers.AnnotatedTable as Ann
 
@@ -216,7 +216,7 @@ cellToJats opts celltype (Ann.Cell (colspec :| _) _colNum cell) =
 
 toAttribs :: Attr -> [Text] -> [(Text, Text)]
 toAttribs (ident, _classes, kvs) knownAttribs =
-  (if T.null ident then id else (("id", ident) :)) $
+  (if T.null ident then id else (("id", escapeNCName ident) :)) $
   filter ((`elem` knownAttribs) . fst) kvs
 
 tableCellToJats :: PandocMonad m