JATS reader: fix handling of alt-text (#9134)

Previously we were looking for an attribute that doesn't exist in JATS; alt-text is provided by a child element. Closes #9130.
jgm · Oct 17, 2023 · fa3513b · fa3513b
1 parent 802c610
commit fa3513b
Show file tree

Hide file tree

Showing 6 changed files with 121 additions and 30 deletions.
diff --git a/src/Text/Pandoc/Readers/JATS.hs b/src/Text/Pandoc/Readers/JATS.hs
@@ -138,18 +138,16 @@ trimNl = T.dropAround (== '\n')
 
 -- function that is used by both graphic (in parseBlock)
 -- and inline-graphic (in parseInline)
-getGraphic :: PandocMonad m
-           => Maybe (Inlines, Text) -> Element -> JATS m Inlines
-getGraphic mbfigdata e = do
+getGraphic :: PandocMonad m => Element -> JATS m Inlines
+getGraphic e = do
   let atVal a = attrValue a e
-      (ident, title, capt) =
-         case mbfigdata of
-           Just (capt', i) -> (i, "fig:" <> atVal "title", capt')
-           Nothing        -> (atVal "id", atVal "title",
-                              text (atVal "alt-text"))
+  let altText = case filterElement (named "alt-text") e of
+         Just alt -> textContent alt
+         Nothing -> mempty
+      (ident, title, altText') = (atVal "id", atVal "title", text altText)
       attr = (ident, T.words $ atVal "role", [])
       imageUrl = atVal "href"
-  return $ imageWith attr imageUrl title capt
+  return $ imageWith attr imageUrl title altText'
 
 getBlocks :: PandocMonad m => Element -> JATS m Blocks
 getBlocks e =  mconcat <$>
@@ -198,7 +196,7 @@ parseBlock (Elem e) = do
         "table-wrap-foot" -> parseBlockWithHeader
         "trans-abstract" -> parseBlockWithHeader
         "verse-group" -> parseBlockWithHeader
-        "graphic" -> para <$> getGraphic Nothing e
+        "graphic" -> para <$> getGraphic e
         "journal-meta" -> parseMetadata e
         "article-meta" -> parseMetadata e
         "custom-meta" -> parseMetadata e
@@ -622,7 +620,7 @@ parseInline (Elem e) =
         "code" -> codeWithLang
         "monospace" -> codeWithLang
 
-        "inline-graphic" -> getGraphic Nothing e
+        "inline-graphic" -> getGraphic e
         "disp-quote" -> do
             qt <- gets jatsQuoteType
             let qt' = if qt == SingleQuote then DoubleQuote else SingleQuote

diff --git a/test/Tests/Readers/JATS.hs b/test/Tests/Readers/JATS.hs
@@ -35,6 +35,42 @@ tests = [ testGroup "inline code"
         , testGroup "images"
           [ test jats "basic" $ "<graphic mimetype=\"image\" mime-subtype=\"\" xlink:href=\"/url\" xlink:title=\"title\" />"
             =?> para (image "/url" "title" mempty)
+            , test jats "alt-text" $ 
+                          "<graphic id=\"graphic001\"\n\
+                          \ xlink:href=\"https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660\"\n\
+                          \ xlink:title=\"This is the title of the graphic\"\n\
+                          \ xlink:role=\"This is the role of the graphic\">\n\
+                          \ <alt-text>Alternative text of the graphic</alt-text>\n\
+                          \ <caption>\n\
+                          \ <title>This is the title of the caption</title>\n\
+                          \ <p>Google doodle from 14 March 2003</p></caption>\n\
+                          \ </graphic>"
+            =?> Para [ Image
+              ( "graphic001"
+              , [ "This"
+                , "is"
+                , "the"
+                , "role"
+                , "of"
+                , "the"
+                , "graphic"
+                ]
+              , []
+              )
+              [ Str "Alternative"
+              , Space
+              , Str "text"
+              , Space
+              , Str "of"
+              , Space
+              , Str "the"
+              , Space
+              , Str "graphic"
+              ]
+              ( "https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
+              , "This is the title of the graphic"
+              )
+              ]
           ]
         , test jats "bullet list" $
                             "<list list-type=\"bullet\">\n\

diff --git a/test/command/5321.md b/test/command/5321.md
@@ -4,7 +4,9 @@
   <caption>
     <p>bar</p>
   </caption>
-  <graphic xlink:href="foo.png" xlink:alt-text="baz" />
+  <graphic xlink:href="foo.png">
+    <alt-text>baz</alt-text>
+  </graphic>
 </fig>
 ^D
 [ Figure
@@ -23,7 +25,9 @@
     <title>foo</title>
     <p>bar</p>
   </caption>
-  <graphic xlink:href="foo.png" xlink:alt-text="baz" />
+    <graphic xlink:href="foo.png">
+      <alt-text>baz</alt-text>
+    </graphic>
 </fig>
 ^D
 [ Figure

diff --git a/test/command/jats-figure-alt-text.md b/test/command/jats-figure-alt-text.md
@@ -5,8 +5,11 @@
     <p>bar</p>
   </caption>
   <alt-text>alternative-decription</alt-text>
-  <graphic xlink:href="foo.png" xlink:alt-text="baz" />
+  <graphic xlink:href="foo.png">
+    <alt-text>baz</alt-text> 
+  </graphic>
 </fig>
+
 ^D
 [ Figure
     ( "fig-1" , [] , [] )
@@ -16,4 +19,4 @@
         [ Image ( "" , [] , [] ) [ Str "baz" ] ( "foo.png" , "" ) ]
     ]
 ]
-```
+```
diff --git a/test/jats-reader.native b/test/jats-reader.native
@@ -745,6 +745,8 @@ Pandoc
   , Para
       [ Str "Einstein"
       , Space
+      , Str "definitely"
+      , Space
       , Str "showed"
       , Space
       , Str "that"
@@ -754,10 +756,29 @@ Pandoc
       [ Para [ Str "Abstract" , Space , Str "text" ]
       , Para
           [ Image
-              ( "" , [] , [] )
-              []
+              ( "graphic001"
+              , [ "This"
+                , "is"
+                , "the"
+                , "role"
+                , "of"
+                , "the"
+                , "graphic"
+                ]
+              , []
+              )
+              [ Str "Alternative"
+              , Space
+              , Str "text"
+              , Space
+              , Str "of"
+              , Space
+              , Str "the"
+              , Space
+              , Str "graphic"
+              ]
               ( "https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
-              , ""
+              , "This is the title of the graphic"
               )
           ]
       , Para [ Math DisplayMath "e=mc^2" ]
@@ -802,11 +823,14 @@ Pandoc
       [ Para [ Str "Abstract" , Space , Str "text" ]
       , Para
           [ Image
-              ( "" , [] , [] )
-              []
-              ( "https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
-              , ""
-              )
+              ( "graphic003" , [] , [] )
+              [ Str "Alternative"
+              , Space
+              , Str "text"
+              , Space
+              , Str "1"
+              ]
+              ( "Title 3" , "" )
           ]
       , Para [ Math DisplayMath "e=mc^2" ]
       ]
@@ -821,6 +845,12 @@ Pandoc
       , Space
       , Str "because\8230"
       ]
+  , Figure
+      ( "fig-1" , [] , [] )
+      (Caption Nothing [ Plain [ Str "bar" ] ])
+      [ Plain [ Str "alternative-decription" ]
+      , Para [ Image ( "" , [] , [] ) [] ( "foo.png" , "" ) ]
+      ]
   , Header
       1
       ( "code-blocks" , [] , [] )

diff --git a/test/jats-reader.xml b/test/jats-reader.xml
@@ -193,19 +193,26 @@
     <title>Formulae with miscelaneus elements</title>
     <p>Regardless of whether or not they are inside a paragraph, should be wrapped in a div, and displayed in a block:</p>
     <p><bold>Inside a paragraph:</bold></p>
-    <p>Einstein showed that 
+    <p>Einstein definitely showed that 
       <disp-formula>
         <abstract>
           <p>Abstract text</p>
         </abstract>
-        <graphic xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660">
-          <alt-text>Alternative text 1</alt-text>
-          <caption><p>Google doodle from 14 March 2003</p></caption>
+        <graphic id="graphic001"
+        xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
+        xlink:title="This is the title of the graphic"
+        xlink:role="This is the role of the graphic">
+          <alt-text>Alternative text of the graphic</alt-text>
+          <caption>
+            <title>This is the title of the caption</title>
+            <p>Google doodle from 14 March 2003</p></caption>
         </graphic>
         <alternatives>
           <tex-math><![CDATA[e=mc^2]]></tex-math>
           <mml:math display="block" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>e</mml:mi><mml:mo>=</mml:mo><mml:mi>m</mml:mi><mml:msup><mml:mi>c</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:math>
-          <graphic xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660">
+          <graphic id="graphic001"
+          xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
+          xlink:title="Title 2">
             <alt-text>Alternative text 2</alt-text>
             <caption><p>Google doodle from 14 March 2003</p></caption>
           </graphic>
@@ -218,14 +225,18 @@
       <abstract>
         <p>Abstract text</p>
       </abstract>
-      <graphic xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660">
+      <graphic id="graphic003"
+      xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
+      xlink:href="Title 3">
         <alt-text>Alternative text 1</alt-text>
         <caption><p>Google doodle from 14 March 2003</p></caption>
       </graphic>
       <alternatives>
         <tex-math><![CDATA[e=mc^2]]></tex-math>
         <mml:math display="block" xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mi>e</mml:mi><mml:mo>=</mml:mo><mml:mi>m</mml:mi><mml:msup><mml:mi>c</mml:mi><mml:mn>2</mml:mn></mml:msup></mml:mrow></mml:math>
-        <graphic xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660">
+        <graphic id="graphic004"
+        xlink:href="https://lh3.googleusercontent.com/dB7iirJ3ncQaVMBGE2YX-WCeoAVIChb6NAzoFcKCFChMsrixJvD7ZRbvcaC-ceXEzXYaoH4K5vaoRDsUyBHFkpIDPnsn3bnzovbvi0a2Gg=s660"
+        xlink:href="Title 4">
           <alt-text>Alternative text 2</alt-text>
           <caption><p>Google doodle from 14 March 2003</p></caption>
         </graphic>
@@ -234,6 +245,15 @@
     <p>This formula is important because…</p>
   </sec>
 </sec>
+<sec id="figures">
+  <fig id="fig-1">
+    <caption>
+      <p>bar</p>
+    </caption>
+    <alt-text>alternative-decription</alt-text>
+    <graphic xlink:href="foo.png" xlink:alt-text="baz" />
+  </fig>
+</sec>
 <sec id="code-blocks">
   <title>Code Blocks</title>
   <p>Code:</p>