Tweak supp file captions (#158)

* Add xsl to introduce label into content surfaced for supp files * Add docs * update kitchen sink fixture to include various iterations of supplementary files * correct filename/heading in docs * update text cases * typos in docs
elifesciences · Jan 17, 2025 · e9e1190 · e9e1190
1 parent f0d638c
commit e9e1190
Show file tree

Hide file tree

Showing 25 changed files with 1,522 additions and 22 deletions.
diff --git a/README.md b/README.md
@@ -205,6 +205,46 @@ The output in Encoda is missing the link:
 
 Since `related-object` can be used in numerous places, this xsl replaces the element with a hyperlink (`<ext-link>`) and if necessary moves it to a different locaiton in the text so that it can be surfaced by EPP.
 
+### [/src/supplementary-materials-label.xsl](/src/supplementary-materials-label.xsl)
+
+This xslt introduces the labels for `<supplementary-material>` into the (title within the) caption. If there is no existing caption with a title, then one is created for the purposes of surfacing the label in EPP. 
+
+This is a workaround until the Encoda JSON output can be improved to surface all content within labels and captions for supplementary material, currently captured in this ticket: https://github.com/elifesciences/enhanced-preprints-issues/issues/1233. Once that ticket is complete, this XSLT can be removed.
+
+At time of writing, for an `object.type == "Link"` EPP will surface the `.content` (and `.target`) in the HTML (`.title` appears to be unhandled, but this where Encoda outputs the label).
+
+As a result of this xslt, the encoda output changes from, for example:
+```json
+{
+  "type": "Link",
+  "target": "supplements/562203_file05.pdf",
+  "title": "This is the label",
+  "content": [
+    "This is the title",
+    "This is the caption"
+  ]
+}
+```
+
+to:
+```json
+{
+  "type": "Link",
+  "target": "supplements/562203_file05.pdf",
+  "title": "This is the label",
+  "content": [
+    {
+      "type": "Strong",
+      "content": [
+        "This is the label. "
+      ]
+    },
+    "This is the title",
+    "This is the caption"
+  ]
+}
+```
+
 # Modify bioRxiv XML in preparation for Encoda
 
 Prerequisites:

diff --git a/src/supplementary-materials-label.xsl b/src/supplementary-materials-label.xsl
@@ -0,0 +1,59 @@
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+    xmlns:xs="http://www.w3.org/2001/XMLSchema"
+    xmlns:xlink="http://www.w3.org/1999/xlink"
+    exclude-result-prefixes="xs"
+    version="3.0">
+
+    <xsl:output method="xml" encoding="UTF-8"/>
+
+     <xsl:template match="*|@*|text()|comment()|processing-instruction()">
+        <xsl:copy>
+            <xsl:apply-templates select="*|@*|text()|comment()|processing-instruction()"/>
+        </xsl:copy>
+    </xsl:template>
+
+    <!-- Introduce a caption so that we can surface the label -->
+    <xsl:template match="supplementary-material[label and not(caption[title or p])]">
+        <xsl:copy>
+            <xsl:apply-templates select="@*|label|label/preceding-sibling::text()"/>
+            <xsl:text>&#xa;</xsl:text>
+            <xsl:element name="caption">
+              <xsl:element name="title">
+                <xsl:element name="bold">
+                  <xsl:apply-templates select="./label[1]/(*|text())"/>
+                </xsl:element>
+              </xsl:element>
+            </xsl:element>
+          <xsl:apply-templates select="*[name()!='label']|text()|comment()|processing-instruction()"/>
+        </xsl:copy>
+    </xsl:template>
+
+    <!-- Introduce the label into the existing caption  -->
+    <xsl:template match="supplementary-material[label and caption[title or p]]">
+      <xsl:copy>
+        <xsl:apply-templates select="@*|label|label/preceding-sibling::text()"/>
+        <xsl:text>&#xa;</xsl:text>
+        <xsl:element name="caption">
+           <xsl:element name="title">
+             <xsl:element name="bold">
+               <xsl:apply-templates select="./label[1]/(*|text())"/>
+               <!-- If it doesn't end with punctuation then add some, otherwise just a space -->
+               <xsl:choose>
+                 <xsl:when test="not(matches([label[1]],'\p{P}$'))">
+                   <xsl:text>. </xsl:text>
+                 </xsl:when>
+                 <xsl:otherwise>
+                   <xsl:text> </xsl:text>
+                 </xsl:otherwise>
+               </xsl:choose>
+             </xsl:element>
+             <xsl:apply-templates select="caption/title/(*|text())"/>
+           </xsl:element>
+          <xsl:apply-templates select="caption/p|caption/title/following-sibling::text()"/>
+        </xsl:element>
+        <xsl:apply-templates select="*[not(name()=('label','caption'))]|caption/following-sibling::text()|comment()|processing-instruction()"/>
+      </xsl:copy>
+    </xsl:template>
+
+
+</xsl:stylesheet>
diff --git a/test/boxed-text-workaround/kitchen-sink.xml b/test/boxed-text-workaround/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/change-label-and-title-elements/kitchen-sink.xml b/test/change-label-and-title-elements/kitchen-sink.xml
@@ -409,9 +409,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/code-workaround/kitchen-sink.xml b/test/code-workaround/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/collate-reference-lists/kitchen-sink.xml b/test/collate-reference-lists/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/conf-ref-workaround/kitchen-sink.xml b/test/conf-ref-workaround/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/convert-doi-links-to-pub-id-elements/kitchen-sink.xml b/test/convert-doi-links-to-pub-id-elements/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/convert-glossaries/kitchen-sink.xml b/test/convert-glossaries/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>

diff --git a/test/disp-quote-workaround/kitchen-sink.xml b/test/disp-quote-workaround/kitchen-sink.xml
@@ -410,9 +410,34 @@ return :shrug:
 <sec sec-type="supplementary-material">
 <title>Supporting information</title>
 <supplementary-material>
-<label>Supplemental figures</label>
+<label>Supp file 1</label>
 <media xlink:href="supplements/468444_file02.pdf"/>
 </supplementary-material>
+<supplementary-material>
+<label>Supp file 2</label>
+<caption>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 3</label>
+<caption>
+<title>This is a title</title>
+<p>This is a para.</p>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<label>Supp file 4.</label>
+<caption>
+<title>This is a title</title>
+</caption>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
+<supplementary-material>
+<media xlink:href="supplements/468444_file03.pdf"/>
+</supplementary-material>
 </sec>
 </body>
 <back>