feat: disallow epub:type on head and metadata content

rdeltour · rdeltour · commit bfb72396ffdf · 2022-12-13T08:42:25.000+01:00
This PR implements the latest post-CR update on where `epub:type` is allowed in HTML content (i.e. everywhere but the `head` element and metadata content). See w3c/epub-specs#2493 Fix #1444, Fix #1445
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-mathml3-inc.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-mathml3-inc.rnc
@@ -3,6 +3,7 @@ namespace local = ""
 namespace x = "http://www.w3.org/1999/xhtml"
 namespace ev = "http://www.w3.org/2001/xml-events"
 namespace ssml = "http://www.w3.org/2001/10/synthesis"
+namespace epub = "http://www.idpf.org/2007/ops"
 
 # #####################################################################
 # 
@@ -19,7 +20,7 @@ include "mathml/mathml3-inc.rnc" {
 
     # extend to circumvent datatype collisions
     NonMathMLAtt =
-        attribute * - (local:* | m:* | xml:* | x:* | ev:* | ssml:*) {
+        attribute * - (local:* | m:* | xml:* | x:* | ev:* | ssml:* | epub:*) {
             datatype.string
         }
 
@@ -30,6 +31,8 @@ include "mathml/mathml3-inc.rnc" {
     annotation-xml = epub.annotation-xml
 }
 # Common attribute extensions
+# - epub:type
+CommonAtt &= epub.type.attr?
 # - SSML attributes 
 CommonAtt &= epub.ssml.ph.attr?
 # - xml:base
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-xhtml-integration.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-xhtml-integration.rnc
@@ -7,89 +7,141 @@
 
 ## combine `epub:type`
 
-  # Allow `epub:type` on the `body` element (even though
-  # it is not defined as palpable content)
-  body.attrs &= epub.type.attr?
-  
-  # Allow `epub:type` on palpable content
+  # Allow `epub:type` on all elements except:
+  #  - the `head` element
+  #  - metadata content (base, link, meta, noscript, script, style, template, title)
   #
   # Note:
   # We can't use a common category since none is defined
-  # for palpable content. So we add it for each element.
-  #
-  # Some elements are only considered palpable under some
-  # conditions; we do not support that at the moment.
-  #
-  a.attrs &= epub.type.attr?
-  abbr.attrs &= epub.type.attr?
-  address.attrs &= epub.type.attr?
-  article.attrs &= epub.type.attr?
-  aside.attrs &= epub.type.attr?
-  audio.attrs &= epub.type.attr?
-  b.attrs &= epub.type.attr?
-  bdi.attrs &= epub.type.attr?
-  bdo.attrs &= epub.type.attr?
-  blockquote.attrs &= epub.type.attr?
-  button.attrs &= epub.type.attr?
+  # for non-metadata content. So we add it for each element.
+
+  # from applications.rnc
+  progress.attrs &= epub.type.attr?
+  dialog.attrs &= epub.type.attr?
+  menu.attrs &= epub.type.attr?
+  mli.attrs &= epub.type.attr?
   canvas.attrs &= epub.type.attr?
-  cite.attrs &= epub.type.attr?
-  code.attrs &= epub.type.attr?
-  data.attrs &= epub.type.attr?
   details.attrs &= epub.type.attr?
-  dfn.attrs &= epub.type.attr?
-  div.attrs &= epub.type.attr?
+  summary.attrs &= epub.type.attr?
+
+  # from block.rnc
+  p.attrs &= epub.type.attr?
+  hr.attrs &= epub.type.attr?
+  pre.attrs &= epub.type.attr?
+  ul.attrs &= epub.type.attr?
+  li.attrs &= epub.type.attr?
+  ol.attrs &= epub.type.attr?
+  oli.attrs &= epub.type.attr?
   dl.attrs &= epub.type.attr?
-  emembed.attrs &= epub.type.attr?
-  fieldset.attrs &= epub.type.attr?
+  dt.attrs &= epub.type.attr?
+  dd.attrs &= epub.type.attr?
+  div.attrs &= epub.type.attr?
+  legend.attrs &= epub.type.attr?
+
+  # from data.rnc
+  time.attrs &= epub.type.attr?
+  time.datetime.attrs &= epub.type.attr?
+  data.attrs &= epub.type.attr?
+  meter.attrs &= epub.type.attr?
+
+  # from embed.rnc
+  img.attrs &= epub.type.attr?
+  picture.attrs &= epub.type.attr?
+  embed.attrs &= epub.type.attr?
+  object.attrs &= epub.type.attr?
+  param.attrs &= epub.type.attr?
+  iframe.attrs &= epub.type.attr?
+  map.attrs &= epub.type.attr?
+  area.attrs &= epub.type.attr?
+
+  # from media.rnc
+  source.attrs &= epub.type.attr?
+  video.attrs &= epub.type.attr?
+  audio.attrs &= epub.type.attr?
+  track.attrs &= epub.type.attr?
   figure.attrs &= epub.type.attr?
-  footer.attrs &= epub.type.attr?
-  form.attrs &= epub.type.attr?
+  figcaption.attrs &= epub.type.attr?
+
+  # from meta.rnc
+  html.attrs &= epub.type.attr?
+  body.attrs &= epub.type.attr?
+
+  # from phrase.rnc
+  a.href.attrs &= epub.type.attr?
+  a.nohref.attrs &= epub.type.attr?
+  em.attrs &= epub.type.attr?
+  strong.attrs &= epub.type.attr?
+  small.attrs &= epub.type.attr?
+  mark.attrs &= epub.type.attr?
+  abbr.attrs &= epub.type.attr?
+  dfn.attrs &= epub.type.attr?
+  i.attrs &= epub.type.attr?
+  b.attrs &= epub.type.attr?
+  s.attrs &= epub.type.attr?
+  u.attrs &= epub.type.attr?
+  code.attrs &= epub.type.attr?
+  var.attrs &= epub.type.attr?
+  samp.attrs &= epub.type.attr?
+  kbd.attrs &= epub.type.attr?
+  sup.attrs &= epub.type.attr?
+  sub.attrs &= epub.type.attr?
+  q.attrs &= epub.type.attr?
+  cite.attrs &= epub.type.attr?
+  span.attrs &= epub.type.attr?
+  bdo.attrs &= epub.type.attr?
+  bdi.attrs &= epub.type.attr?
+  br.attrs &= epub.type.attr?
+  wbr.attrs &= epub.type.attr?
+
+  # from revision.rnc
+  ins.attrs &= epub.type.attr?
+  del.attrs &= epub.type.attr?
+
+  # from ruby.rnc
+  ruby.attrs &= epub.type.attr?
+  rt.attrs &= epub.type.attr?
+  rp.attrs &= epub.type.attr?
+
+  # from sectional.rnc
   h1.attrs &= epub.type.attr?
   h2.attrs &= epub.type.attr?
   h3.attrs &= epub.type.attr?
   h4.attrs &= epub.type.attr?
   h5.attrs &= epub.type.attr?
   h6.attrs &= epub.type.attr?
-  header.attrs &= epub.type.attr?
   hgroup.attrs &= epub.type.attr?
-  i.attrs &= epub.type.attr?
-  iframe.attrs &= epub.type.attr?
-  img.attrs &= epub.type.attr?
-  input.attrs &= epub.type.attr?
-  ins.attrs &= epub.type.attr?
-  kbd.attrs &= epub.type.attr?
-  label.attrs &= epub.type.attr?
-  main.attrs &= epub.type.attr?
-  map.attrs &= epub.type.attr?
-  mark.attrs &= epub.type.attr?
-  menu.attrs &= epub.type.attr?
-  meter.attrs &= epub.type.attr?
-  nav.attrs &= epub.type.attr?
-  object.attrs &= epub.type.attr?
-  ol.attrs &= epub.type.attr?
-  output.attrs &= epub.type.attr?
-  p.attrs &= epub.type.attr?
-  pre.attrs &= epub.type.attr?
-  progress.attrs &= epub.type.attr?
-  q.attrs &= epub.type.attr?
-  ruby.attrs &= epub.type.attr?
-  s.attrs &= epub.type.attr?
-  samp.attrs &= epub.type.attr?
+  address.attrs &= epub.type.attr?
+  blockquote.attrs &= epub.type.attr?
+
+  # from structural.rnc
   section.attrs &= epub.type.attr?
-  select.attrs &= epub.type.attr?
-  small.attrs &= epub.type.attr?
-  span.attrs &= epub.type.attr?
-  strong.attrs &= epub.type.attr?
-  sub.attrs &= epub.type.attr?
-  sup.attrs &= epub.type.attr?
+  nav.attrs &= epub.type.attr?
+  article.attrs &= epub.type.attr?
+  aside.attrs &= epub.type.attr?
+  header.attrs &= epub.type.attr?
+  footer.attrs &= epub.type.attr?
+  main.attrs &= epub.type.attr?
+
+  # from tables.rnc
   table.attrs &= epub.type.attr?
-  text.attrs &= epub.type.attr?
-  area.attrs &= epub.type.attr?
-  time.attrs &= epub.type.attr?
-  u.attrs &= epub.type.attr?
-  ul.attrs &= epub.type.attr?
-  var.attrs &= epub.type.attr?
-  video.attrs &= epub.type.attr?
+  caption.attrs &= epub.type.attr?
+  colgroup.attrs &= epub.type.attr?
+  thead.attrs &= epub.type.attr?
+  tfoot.attrs &= epub.type.attr?
+  tbody.attrs &= epub.type.attr?
+  tr.attrs &= epub.type.attr?
+  td.attrs &= epub.type.attr?
+  th.attrs &= epub.type.attr?
+
+  # from web-forms.rnc and web-forms2.rnc
+  common-form.attrs &= epub.type.attr?
+  option.attrs &= epub.type.attr?
+  optgroup.attrs &= epub.type.attr?
+  form.attrs &= epub.type.attr?
+  label.attrs &= epub.type.attr?
+  output.attrs &= epub.type.attr?
+  datalist.attrs &= epub.type.attr?
+
 
 ## combine prefix attributes
 
diff --git a/src/test/resources/epub-region-nav/region-nav-publication.feature b/src/test/resources/epub-region-nav/region-nav-publication.feature
@@ -91,5 +91,4 @@ Feature: EPUB Region-Based Navigation ▸ Full Publication Checks
 
   Scenario: Verify subregion navigation using comics semantics
     When checking EPUB 'region-based-nav-comics-valid'
-    Then error RSC-005 is reported 4 times (non-palpable content)
     Then no errors or warnings are reported
diff --git a/src/test/resources/epub3/06-content-document/content-document-xhtml.feature b/src/test/resources/epub3/06-content-document/content-document-xhtml.feature
@@ -645,14 +645,15 @@ Feature: EPUB 3 — Content Documents — XHTML
 
   #### 6.1.3.1 Structural semantics
 
-  Scenario: Verify `epub:type` attribute with valid semantic
+  @spec @xref:sec-xhtml-structural-semantics
+  Scenario: Verify `epub:type` attribute on allowed content
     When checking document 'epubtype-valid.xhtml'
     Then no errors or warnings are reported
 
   @spec @xref:sec-xhtml-structural-semantics
-  Scenario: Report `epub:type` attribute on non-palpable content
-    When checking document 'epubtype-on-non-palpable-content-error.xhtml'
-    Then error RSC-005 is reported 2 times
+  Scenario: Report `epub:type` attribute on 'head' or metadata content 
+    When checking document 'epubtype-disallowed-error.xhtml'
+    Then error RSC-005 is reported 8 times
     And no other errors or warnings are reported
   
   Scenario: Verify `epub:type` attribute with reserved vocabulary
@@ -678,8 +679,7 @@ Feature: EPUB 3 — Content Documents — XHTML
 
   Scenario: Verify `epub:type` attribute that does not follow usage suggestions
     Given the reporting level set to usage
-    When checking document 'epubtype-disallowed-usage.xhtml'
-    Then error RSC-005 is reported 3 times (non-palpable content)
+    When checking document 'epubtype-misuse-usage.xhtml'
     Then usage OPF-087 is reported 7 times
     And no other errors or warnings are reported
 
diff --git a/src/test/resources/epub3/06-content-document/files/epubtype-disallowed-error.xhtml b/src/test/resources/epub3/06-content-document/files/epubtype-disallowed-error.xhtml
@@ -0,0 +1,17 @@
+<!DOCTYPE html>	 
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops"
+	epub:prefix="test: https://example.org/vocab/#" xml:lang="en" lang="en">
+	<head epub:type="test:invalid">
+		<meta epub:type="test:invalid" charset="utf-8" />
+		<title epub:type="test:invalid">epub:type in head element</title>
+		<meta epub:type="test:invalid" name="creator" content="Herman Melville"  />
+		<style epub:type="test:invalid"></style>
+		<link epub:type="test:invalid" type="text/css" rel="stylesheet" href="style.css" />
+	</head>
+	<body>
+		<script epub:type="test:invalid" src="script.js"></script>
+		<noscript epub:type="test:invalid">text</noscript>
+		<h1>Test</h1>
+		<div></div>
+	</body>
+</html>
diff --git a/src/test/resources/epub3/06-content-document/files/epubtype-misuse-usage.xhtml b/src/test/resources/epub3/06-content-document/files/epubtype-misuse-usage.xhtml
diff --git a/src/test/resources/epub3/06-content-document/files/epubtype-on-non-palpable-content-error.xhtml b/src/test/resources/epub3/06-content-document/files/epubtype-on-non-palpable-content-error.xhtml
diff --git a/src/test/resources/epub3/06-content-document/files/epubtype-valid.xhtml b/src/test/resources/epub3/06-content-document/files/epubtype-valid.xhtml
@@ -12,5 +12,7 @@
 			<h2>endnotes</h2>
 			<div epub:type="endnote">endnote</div>
 		</section>
+		<a href="link.xhtml" epub:type="tip">link</a>
+		<a epub:type="glossref">anchor</a>
 	</body>
 </html>
