feat: restrict SVG title element to HTML elements

This commit updates the content model of the SVG title element to allow any text or HTML elements, but reject any non-HTML element. - use NVDL to reject non-HTML namespaces - the master XHTML schema is now the NVDL schema - add an element class to include all HTML elements Fix #1342
w3c · Nov 27, 2022 · 8631b7d · 8631b7d
1 parent 3086258
commit 8631b7d
Show file tree

Hide file tree

Showing 26 changed files with 224 additions and 107 deletions.
diff --git a/src/main/java/com/adobe/epubcheck/ops/OPSChecker.java b/src/main/java/com/adobe/epubcheck/ops/OPSChecker.java
@@ -56,7 +56,7 @@ public class OPSChecker extends PublicationResourceChecker
       .putAll(Predicates.and(mimetype("application/xhtml+xml"), version(EPUBVersion.VERSION_2)),
           XMLValidators.XHTML_20_NVDL, XMLValidators.XHTML_20_SCH, XMLValidators.IDUNIQUE_20_SCH)
       .putAll(Predicates.and(mimetype("application/xhtml+xml"), version(EPUBVersion.VERSION_3)),
-          XMLValidators.XHTML_30_RNC, XMLValidators.XHTML_30_SCH, XMLValidators.SVG_INFORMATIVE_30_NVDL)
+          XMLValidators.XHTML_30_NVDL, XMLValidators.SVG_INFORMATIVE_30_NVDL)
       .putAll(Predicates.and(mimetype("image/svg+xml"), version(EPUBVersion.VERSION_2)),
           XMLValidators.SVG_20_NVDL, XMLValidators.IDUNIQUE_20_SCH)
       .putAll(Predicates.and(mimetype("image/svg+xml"), version(EPUBVersion.VERSION_3)),

diff --git a/src/main/java/com/adobe/epubcheck/xml/XMLValidators.java b/src/main/java/com/adobe/epubcheck/xml/XMLValidators.java
@@ -45,6 +45,7 @@ public enum XMLValidators
   XHTML_20_SCH("schema/20/sch/xhtml.sch"),
   XHTML_30_SCH("schema/30/epub-xhtml-30.sch"),
   XHTML_30_RNC("schema/30/epub-xhtml-30.rnc"),
+  XHTML_30_NVDL("schema/30/epub-xhtml-30.nvdl"),
   XHTML_EDUPUB_STRUCTURE_SCH("schema/30/edupub/edu-structure.sch"),
   XHTML_EDUPUB_SEMANTICS_SCH("schema/30/edupub/edu-semantics.sch"),
   XHTML_DATANAV_SCH("schema/30/datanav/datanav-xhtml.sch"),

diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-30.nvdl b/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-30.nvdl
@@ -5,6 +5,7 @@
 			<validate schema="epub-svg-30.rnc" schemaType="application/relax-ng-compact-syntax"
 				useMode="allowForeignNS">
 				<context path="foreignObject" useMode="attachAnyNS"/>
+				<context path="title" useMode="allowOnlyHTML"/>
 			</validate>
 			<validate schema="epub-svg-30.sch" useMode="attachAnyNS"/>
 		</namespace>
@@ -34,4 +35,27 @@
 			<attach/>
 		</anyNamespace>
 	</mode>
+	<mode name="allowOnlyHTML">
+		<namespace ns="http://www.w3.org/1999/xhtml">
+			<attach/>
+		</namespace>
+		<namespace ns="http://www.idpf.org/2007/ops" match="attributes">
+			<attach/>
+		</namespace>
+		<namespace ns="http://www.w3.org/1999/xlink" match="attributes">
+			<attach/>
+		</namespace>
+		<namespace ns="http://www.w3.org/XML/1998/namespace" match="attributes">
+			<attach/>
+		</namespace>
+		<namespace ns="" match="attributes">
+			<attach/>
+		</namespace>
+		<anyNamespace match="attributes">
+			<allow/>
+		</anyNamespace>
+		<anyNamespace>
+			<reject/>
+		</anyNamespace>
+	</mode>
 </rules>
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-30.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-30.rnc
@@ -1,17 +1,17 @@
 namespace svg = "http://www.w3.org/2000/svg"
 
-include "./mod/epub-xhtml.rnc" {
+include "./mod/epub-xhtml-inc.rnc" {
   start = svg
 }
 include "./mod/epub-mathml3.rnc"
 include "./mod/epub-svg-inc.rnc"
 
-svg.title.content |= common.elem.phrasing
+svg.title.content |= common.inner.anyhtml
 
 svg.foreignObject.content |= 
   ( body.elem
   | common.inner.flow
   | math 
   )
 
-mtext.content |= common.elem.phrasing
+mtext.content |= common.elem.phrasing
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-informative.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/epub-svg-informative.rnc
@@ -17,14 +17,5 @@ SVG.Core.attrib &= aria.global?
 SVG.Core.extra.attrib &= epub.type.attr?
 
 ## Do not check restricted elements, they are checked in normative schemas
-SVG.foreignObject.content |= any.content
-SVG.title.content |= any.content
-
-## Anything
-
-## Any attribute from any namespace, other than the special cases
-any.attr = attribute * { text }*
-## Any element from any namespace, other than the special cases
-any.elem = element * { any.content & any.attr }
-## Any content from any namespace
-any.content = text & any.elem*
+SVG.foreignObject.content |= common.inner.anything
+SVG.title.content |= common.inner.anything 
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.nvdl b/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.nvdl
@@ -1,15 +1,46 @@
 <?xml version="1.0" encoding="UTF-8"?>
-<rules xmlns="http://purl.oclc.org/dsdl/nvdl/ns/structure/1.0" startMode="html">
+<rules xmlns="http://purl.oclc.org/dsdl/nvdl/ns/structure/1.0"
+  xmlns:svg="http://www.w3.org/2000/svg" startMode="html">
   <mode name="html">
     <namespace ns="http://www.w3.org/1999/xhtml">
       <validate schema="epub-xhtml-30.rnc" schemaType="application/relax-ng-compact-syntax"
-        useMode="attach"/>
+        useMode="attach">
+        <context path="title" useMode="allowOnlyHTML"/>
+      </validate>
       <validate schema="epub-xhtml-30.sch" useMode="attach"/>
     </namespace>
   </mode>
   <mode name="attach">
+    <namespace ns="http://www.w3.org/2000/svg">
+      <attach>
+        <context path="title" useMode="allowOnlyHTML"/>
+      </attach>
+    </namespace>
     <anyNamespace>
       <attach/>
     </anyNamespace>
   </mode>
+  <mode name="allowOnlyHTML">
+    <namespace ns="http://www.w3.org/1999/xhtml">
+      <attach/>
+    </namespace>
+    <namespace ns="http://www.idpf.org/2007/ops" match="attributes">
+      <attach/>
+    </namespace>
+    <namespace ns="http://www.w3.org/1999/xlink" match="attributes">
+      <attach/>
+    </namespace>
+    <namespace ns="http://www.w3.org/XML/1998/namespace" match="attributes">
+      <attach/>
+    </namespace>
+    <namespace ns="" match="attributes">
+      <attach/>
+    </namespace>
+    <anyNamespace match="attributes">
+      <allow/>
+    </anyNamespace>
+    <anyNamespace>
+      <reject/>
+    </anyNamespace>
+  </mode>
 </rules>
diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.rnc
@@ -1,6 +1,6 @@
 namespace svg = "http://www.w3.org/2000/svg"
 
-include "./mod/epub-xhtml.rnc"
+include "./mod/epub-xhtml-inc.rnc"
 include "./mod/epub-mathml3.rnc"
 include "./mod/epub-svg-inc.rnc" {
   svg.attr.id = attribute id { datatype.html5.token }?
@@ -11,7 +11,7 @@ common.elem.phrasing |= math
 
 math.attributes &= aria.global?
 
-svg.title.content |= common.elem.phrasing
+svg.title.content |= common.inner.anyhtml
 
 svg.foreignObject.content |= 
   ( common.inner.flow

diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.sch b/src/main/resources/com/adobe/epubcheck/schema/30/epub-xhtml-30.sch
@@ -152,11 +152,6 @@
         <param name="descendant" value="h:label"/>
     </pattern>
 
-    <pattern id="descendant-svgtitle-svg" is-a="disallowed-descendants">
-        <param name="ancestor" value="svg:title"/>
-        <param name="descendant" value="svg:*"/>
-    </pattern>
-
     <pattern id="bdo-dir" is-a="required-attr">
         <param name="elem" value="h:bdo"/>
         <param name="attr" value="dir"/>

diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-svg-inc.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-svg-inc.rnc
@@ -27,7 +27,7 @@ svg.foreignObject.content = notAllowed
 ### SVG title element restrictions
 svg.title =
   element title { any.attr & svg.title.content }
-svg.title.content = text
+svg.title.content = text 
 
 ## Anything
 

diff --git a/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-xhtml-inc.rnc b/src/main/resources/com/adobe/epubcheck/schema/30/mod/epub-xhtml-inc.rnc
@@ -0,0 +1,18 @@
+
+# #####################################################################
+##  RELAX NG Schema for EPUB HTML 5: For inclusion in master schema   #
+# #####################################################################
+
+## Include the main XHTML schema
+
+include "./epub-xhtml.rnc"
+
+## HTML fragments
+
+# Note: the default HMTL schema does not define such a category, so we
+# do our best to include what makes sense here
+
+  common.inner.anyhtml = 
+      html.elem
+    | body.elem
+    | common.inner.flow
diff --git a/src/test/resources/epub3/00-minimal/files/minimal/EPUB/content_001.xhtml b/src/test/resources/epub3/00-minimal/files/minimal/EPUB/content_001.xhtml
@@ -1,6 +1,6 @@
 <!DOCTYPE html>
-<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
-	<head>
+<html xmlns:epub="http://www.idpf.org/2007/ops" xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+	<head epub:type="document">
 		<meta charset="utf-8"/>
 		<title>Minimal EPUB</title>
 	</head>

diff --git a/src/test/resources/epub3/06-content-document/content-document-svg.feature b/src/test/resources/epub3/06-content-document/content-document-svg.feature
@@ -101,6 +101,8 @@ Feature: EPUB 3 — Content Documents — SVG
 
   ###  6.2.3 Restrictions on SVG
 
+	#### `foreignObject` element
+
   @spec @xref:sec-svg-restrictions
   Scenario: Verify that `foreignObject` conforming to the rules is allowed
     When checking document 'foreignObject-valid.svg'
@@ -132,32 +134,35 @@ Feature: EPUB 3 — Content Documents — SVG
     And the message contains 'element "body" not allowed here'
     And no other errors or warnings are reported
 
+  @spec @xref:sec-svg-restrictions
   Scenario: Report HTML validation errors within `foreignObject` content
     When checking document 'foreignObject-html-invalid-error.svg'
     Then error RSC-005 is reported
     And the message contains 'attribute "href" not allowed here'
     And no other errors or warnings are reported
 
-  @spec @xref:sec-svg-restrictions
-  Scenario: Verify `title` can contain text
-    When checking document 'title-text-valid.svg'
-    Then no errors or warnings are reported
+	#### `title` element
 
   @spec @xref:sec-svg-restrictions
-  Scenario: Verify `title` can contain HTML phrasing content
-    When checking document 'title-phrasing-content-valid.svg'
+  Scenario: Verify `title` valid content model
+    When checking document 'title-content-valid.svg'
     Then no errors or warnings are reported
 
   @spec @xref:sec-svg-restrictions
-  Scenario: Report `title` with non-phrasing content
-    When checking document 'title-not-phrasing-content-error.svg'
+  Scenario: Report `title` with non-HTML elements
+    When checking document 'title-content-not-html-error.svg'
+    Then error RSC-005 is reported
+    And the message contains 'elements from namespace "https://example.org" are not allowed'
     Then error RSC-005 is reported
-    And the message contains 'element "h1" not allowed here'
+    And the message contains 'elements from namespace "http://www.w3.org/2000/svg" are not allowed'
     And no other errors or warnings are reported
 
+  @spec @xref:sec-svg-restrictions
   Scenario: Report HTML validation errors within `title` content
-    When checking document 'title-html-invalid-error.svg'
+    When checking document 'title-content-invalid-html-error.svg'
     Then error RSC-005 is reported
     And the message contains 'attribute "href" not allowed here'
+    Then error RSC-005 is reported
+    And the message contains 'element "body" not allowed here'
     And no other 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
@@ -985,22 +985,20 @@ Feature: EPUB 3 — Content Documents — XHTML
     And the message contains 'element "title" not allowed here'
     And no other errors or warnings are reported
 
-  Scenario: Verify `title` can contain text
-    When checking document 'svg-title-text-valid.xhtml'
+  Scenario: Verify `title` valid content model
+    When checking document 'svg-title-content-valid.xhtml'
     Then no errors or warnings are reported
 
-  Scenario: Verify `title` can contain HTML phrasing content
-    When checking document 'svg-title-phrasing-content-valid.xhtml'
-    Then no errors or warnings are reported
-
-  Scenario: Report `title` with non-phrasing content
-    When checking document 'svg-title-not-phrasing-content-error.xhtml'
+  Scenario: Report `title` with non-HTML elements
+    When checking document 'svg-title-content-not-html-error.xhtml'
+    Then error RSC-005 is reported
+    And the message contains 'elements from namespace "https://example.org" are not allowed'
     Then error RSC-005 is reported
-    And the message contains 'element "h1" not allowed here'
+    And the message contains 'elements from namespace "http://www.w3.org/2000/svg" are not allowed'
     And no other errors or warnings are reported
 
   Scenario: Report HTML validation errors within `title` content
-    When checking document 'svg-title-html-invalid-error.xhtml'
+    When checking document 'svg-title-content-invalid-html-error.xhtml'
     Then error RSC-005 is reported
     And the message contains 'attribute "href" not allowed here'
     And no other errors or warnings are reported

diff --git a/src/test/resources/epub3/06-content-document/files/epubtype-not-allowed-error.svg b/src/test/resources/epub3/06-content-document/files/epubtype-not-allowed-error.svg
diff --git a/.../files/svg-title-html-invalid-error.xhtml → ...vg-title-content-invalid-html-error.xhtml b/.../files/svg-title-html-invalid-error.xhtml → ...vg-title-content-invalid-html-error.xhtml
diff --git a/src/test/resources/epub3/06-content-document/files/svg-title-content-not-html-error.xhtml b/src/test/resources/epub3/06-content-document/files/svg-title-content-not-html-error.xhtml
@@ -0,0 +1,26 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8" />
+		<title>Test</title>
+	</head>
+	<body>
+		<h1>Test</h1>
+		<svg width="12cm" height="4cm" viewBox="0 0 1200 400" version="1.1"
+			xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+			<title>Test</title>
+			<desc>Non-HTML content</desc>
+			<g>
+				<title><not:html xmlns:not="https://example.org">not HTML</not:html></title>
+			</g>
+			<g>
+				<title><body xmlns="http://www.w3.org/1999/xhtml">
+						<svg xmlns="http://www.w3.org/2000/svg">
+							<title>Inner SVG</title>
+							<desc>Disallowed</desc>
+						</svg>
+					</body></title>
+			</g>
+		</svg>
+	</body>
+</html>
diff --git a/src/test/resources/epub3/06-content-document/files/svg-title-content-valid.xhtml b/src/test/resources/epub3/06-content-document/files/svg-title-content-valid.xhtml
@@ -0,0 +1,35 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8" />
+		<title>Test</title>
+	</head>
+	<body>
+		<h1>Test</h1>
+		<svg width="12cm" height="4cm" viewBox="0 0 1200 400" version="1.1"
+			xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+			<title>title</title>
+			<desc>Example</desc>
+			<g>
+				<title>Plain text</title>
+			</g>
+			<g>
+				<title><span xmlns="http://www.w3.org/1999/xhtml">html phrasing content</span></title>
+			</g>
+			<g>
+				<title><h1 xmlns="http://www.w3.org/1999/xhtml">html flow content</h1></title>
+			</g>
+			<g>
+				<title><body xmlns="http://www.w3.org/1999/xhtml">html non-categorized elements</body></title>
+			</g>
+			<g>
+				<title>some <span xmlns="http://www.w3.org/1999/xhtml">mixed</span> content</title>
+			</g>
+			<g>
+				<title><html xmlns="http://www.w3.org/1999/xhtml" lang="en">
+					<head><title>full html document</title></head><body>…</body></html>
+				</title>
+			</g>
+		</svg>
+	</body>
+</html>
diff --git a/...test/resources/epub3/06-content-document/files/svg-title-not-phrasing-content-error.xhtml b/...test/resources/epub3/06-content-document/files/svg-title-not-phrasing-content-error.xhtml
diff --git a/src/test/resources/epub3/06-content-document/files/svg-title-phrasing-content-valid.xhtml b/src/test/resources/epub3/06-content-document/files/svg-title-phrasing-content-valid.xhtml