feat: allow SVG/MathML doctype declarations

EPUB 3.3. now allows a reserved set of external identifiers in doctype declarations of documents with select media types. See: https://www.w3.org/TR/epub-33/#app-identifiers-allowed This commit: - adds those as special cases to the XML parser code - totally removes entity fetching for EPUB 3.3 - keeps forbidding external entities in the internal subset Fix #1192, Fix #1114
w3c · Jan 23, 2022 · 6e44b39 · 6e44b39
1 parent ab99f1d
commit 6e44b39
Show file tree

Hide file tree

Showing 23 changed files with 215 additions and 10 deletions.
diff --git a/src/main/java/com/adobe/epubcheck/xml/XMLParser.java b/src/main/java/com/adobe/epubcheck/xml/XMLParser.java
@@ -313,20 +313,18 @@ public InputSource resolveEntity(String publicId, String systemId)
 
     String resourcePath = systemIdMap.get(systemId);
 
-    if (resourcePath != null)
+    // external entities are not resolved in EPUB 3
+    if (context.version == EPUBVersion.VERSION_3 || systemId.equals("about:legacy-compat")) {
+      return new InputSource(new StringReader(""));
+    }
+    else if (resourcePath != null)
     {
       InputStream resourceStream = ResourceUtil.getResourceStream(resourcePath);
       InputSource source = new InputSource(resourceStream);
       source.setPublicId(publicId);
       source.setSystemId(systemId);
       return source;
     }
-    else if (systemId.equals("about:legacy-compat"))
-    {
-      // special case
-      return new InputSource(new StringReader(""));
-
-    }
     else
     {
       // check for a system prop that turns off online fetching
@@ -797,7 +795,24 @@ else if (context.version == EPUBVersion.VERSION_3)
       }
       else if (publicId != null || systemId != null)
       {
-        report.message(MessageId.OPF_073, getLocation());
+        // check if the declaration is allowed for the current media type
+        boolean isAllowed;
+        switch (mimeType)
+        {
+        case "image/svg+xml":
+          isAllowed = "-//W3C//DTD SVG 1.1//EN".equals(publicId) && "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd".equals(systemId);
+          break;
+        case "application/mathml+xml":
+        case "application/mathml-content+xml":
+        case "application/mathml-presentation+xml":
+          isAllowed = "-//W3C//DTD MathML 3.0//EN".equals(publicId) && "http://www.w3.org/Math/DTD/mathml3/mathml3.dtd".equals(systemId);
+          break;
+        default:
+          isAllowed= false;
+        }
+        if (!isAllowed) {
+          report.message(MessageId.OPF_073, getLocation());
+        }
       }
     }
 

diff --git a/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/content_001.xhtml b/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/content_001.xhtml
@@ -0,0 +1,11 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8"/>
+		<title>Minimal EPUB</title>
+	</head>
+	<body>
+		<h1>Loomings</h1>
+		<p>Call me Ishmael.</p>
+	</body>
+</html>
diff --git a/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-1.xml b/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-1.xml
@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE math PUBLIC "-//W3C//DTD MathML 3.0//EN" "http://www.w3.org/Math/DTD/mathml3/mathml3.dtd">
+<math xmlns="http://www.w3.org/1998/Math/MathML" alttext="2x+y-z">
+  <mrow>
+    <mn>2</mn>
+    <mo> &#x2061;<!--INVISIBLE TIMES--></mo>
+    <mi>x</mi>
+  </mrow>
+  <mrow>
+    <mo>+</mo>
+    <mi>y</mi>
+    <mo>-</mo>
+    <mi>z</mi>
+  </mrow>
+</math>
diff --git a/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-2.xml b/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-2.xml
@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE math PUBLIC "-//W3C//DTD MathML 3.0//EN" "http://www.w3.org/Math/DTD/mathml3/mathml3.dtd">
+<math xmlns="http://www.w3.org/1998/Math/MathML" alttext="2x+y-z">
+  <mrow>
+    <mn>2</mn>
+    <mo> &#x2061;<!--INVISIBLE TIMES--></mo>
+    <mi>x</mi>
+  </mrow>
+  <mrow>
+    <mo>+</mo>
+    <mi>y</mi>
+    <mo>-</mo>
+    <mi>z</mi>
+  </mrow>
+</math>
diff --git a/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-3.xml b/...ources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/mathml-mediatype-3.xml
@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE math PUBLIC "-//W3C//DTD MathML 3.0//EN" "http://www.w3.org/Math/DTD/mathml3/mathml3.dtd">
+<math xmlns="http://www.w3.org/1998/Math/MathML" alttext="2x+y-z">
+  <mrow>
+    <mn>2</mn>
+    <mo> &#x2061;<!--INVISIBLE TIMES--></mo>
+    <mi>x</mi>
+  </mrow>
+  <mrow>
+    <mo>+</mo>
+    <mi>y</mi>
+    <mo>-</mo>
+    <mi>z</mi>
+  </mrow>
+</math>
diff --git a/...-external-identifier-error/EPUB/nav.xhtml → ...l-identifier-allowed-valid/EPUB/nav.xhtml b/...-external-identifier-error/EPUB/nav.xhtml → ...l-identifier-allowed-valid/EPUB/nav.xhtml
diff --git a/src/test/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/package.opf b/src/test/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/package.opf
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<package xmlns="http://www.idpf.org/2007/opf" version="3.0" xml:lang="en" unique-identifier="q">
+<metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
+  <dc:title id="title">Minimal EPUB 3.0</dc:title>
+  <dc:language>en</dc:language>
+  <dc:identifier id="q">NOID</dc:identifier>
+  <meta property="dcterms:modified">2017-06-14T00:00:01Z</meta>
+</metadata>
+<manifest>
+  <item id="content_001"  href="content_001.xhtml" media-type="application/xhtml+xml"/>
+  <item id="nav"  href="nav.xhtml" media-type="application/xhtml+xml" properties="nav"/>
+  <item id="mathml-1" href="mathml-mediatype-1.xml" media-type="application/mathml+xml" fallback="content_001"/>
+  <item id="mathml-2" href="mathml-mediatype-2.xml" media-type="application/mathml-presentation+xml" fallback="content_001"/>
+  <item id="mathml-3" href="mathml-mediatype-3.xml" media-type="application/mathml-content+xml" fallback="content_001"/>
+  <item id="svg" href="svg.svg" media-type="image/svg+xml"/>
+</manifest>
+<spine>
+  <itemref idref="content_001" />
+  <itemref idref="svg"/>
+  <itemref idref="mathml-1"/>
+  <itemref idref="mathml-2"/>
+  <itemref idref="mathml-3"/>
+</spine>
+</package>
diff --git a/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/svg.svg b/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/EPUB/svg.svg
diff --git a/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/META-INF/container.xml b/...t/resources/epub3/files/epub/xml-external-identifier-allowed-valid/META-INF/container.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8" ?>
+<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
+	<rootfiles>
+		<rootfile full-path="EPUB/package.opf" media-type="application/oebps-package+xml"/>
+	</rootfiles>
+</container>
diff --git a/...octype-external-identifier-error/mimetype → ...xternal-identifier-allowed-valid/mimetype b/...octype-external-identifier-error/mimetype → ...xternal-identifier-allowed-valid/mimetype
diff --git a/...urces/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/content_001.xhtml b/...urces/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/content_001.xhtml
@@ -0,0 +1,11 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8"/>
+		<title>Minimal EPUB</title>
+	</head>
+	<body>
+		<h1 id="ch1">Loomings</h1>
+		<p>Call me Ishmael.</p>
+	</body>
+</html>
diff --git a/...est/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/nav.xhtml b/...est/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/nav.xhtml
@@ -0,0 +1,14 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8"/>
+		<title>Minimal Nav</title>
+	</head>
+	<body>
+		<nav epub:type="toc">
+			<ol>
+				<li><a href="content_001.xhtml">content 001</a></li>
+			</ol>
+		</nav>
+	</body>
+</html>
diff --git a/...xternal-identifier-error/EPUB/package.opf → ...fier-bad-mediatype-error/EPUB/package.opf b/...xternal-identifier-error/EPUB/package.opf → ...fier-bad-mediatype-error/EPUB/package.opf
diff --git a/src/test/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/toc.ncx b/src/test/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/EPUB/toc.ncx
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<ncx xmlns:ncx="http://www.daisy.org/z3986/2005/ncx/"
+    xmlns="http://www.daisy.org/z3986/2005/ncx/"
+    version="2005-1"
+    xml:lang="en">
+    <head>
+        <meta name="dtb:uid" content="NOID"/>
+        <meta name="dtb:depth" content="1"/>
+        <meta name="dtb:totalPageCount" content="0"/>
+        <meta name="dtb:maxPageNumber" content="0"/>
+    </head>
+    <docTitle>
+        <text>NCX</text>
+    </docTitle>
+    <navMap>
+        <navPoint id="ch1" playOrder="1">
+            <navLabel>
+                <text>Chapter 1</text>
+            </navLabel>
+            <content src="content_001.xhtml#ch1"/>
+        </navPoint>
+    </navMap>
+</ncx>
diff --git a/...urces/epub3/files/epub/xml-external-identifier-bad-mediatype-error/META-INF/container.xml b/...urces/epub3/files/epub/xml-external-identifier-bad-mediatype-error/META-INF/container.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8" ?>
+<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
+	<rootfiles>
+		<rootfile full-path="EPUB/package.opf" media-type="application/oebps-package+xml"/>
+	</rootfiles>
+</container>
diff --git a/src/test/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/mimetype b/src/test/resources/epub3/files/epub/xml-external-identifier-bad-mediatype-error/mimetype
@@ -0,0 +1 @@
+application/epub+zip
diff --git a/...l-identifier-error/EPUB/content_001.xhtml → ...r-disallowed-error/EPUB/content_001.xhtml b/...l-identifier-error/EPUB/content_001.xhtml → ...r-disallowed-error/EPUB/content_001.xhtml
diff --git a/src/test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/EPUB/nav.xhtml b/src/test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/EPUB/nav.xhtml
@@ -0,0 +1,14 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xml:lang="en" lang="en">
+	<head>
+		<meta charset="utf-8"/>
+		<title>Minimal Nav</title>
+	</head>
+	<body>
+		<nav epub:type="toc">
+			<ol>
+				<li><a href="content_001.xhtml">content 001</a></li>
+			</ol>
+		</nav>
+	</body>
+</html>
diff --git a/...test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/EPUB/package.opf b/...test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/EPUB/package.opf
@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<package xmlns="http://www.idpf.org/2007/opf" version="3.0" xml:lang="en" unique-identifier="q">
+<metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
+  <dc:title id="title">Minimal EPUB 3.0</dc:title>
+  <dc:language>en</dc:language>
+  <dc:identifier id="q">NOID</dc:identifier>
+  <meta property="dcterms:modified">2017-06-14T00:00:01Z</meta>
+</metadata>
+<manifest>
+  <item id="content_001"  href="content_001.xhtml" media-type="application/xhtml+xml"/>
+  <item id="nav"  href="nav.xhtml" media-type="application/xhtml+xml" properties="nav"/>
+  <item id="ncx" href="toc.ncx" media-type="application/x-dtbncx+xml" />
+</manifest>
+<spine toc="ncx">
+  <itemref idref="content_001" />
+</spine>
+</package>
diff --git a/...pe-external-identifier-error/EPUB/toc.ncx → ...-identifier-disallowed-error/EPUB/toc.ncx b/...pe-external-identifier-error/EPUB/toc.ncx → ...-identifier-disallowed-error/EPUB/toc.ncx
diff --git a/...l-identifier-error/META-INF/container.xml → ...r-disallowed-error/META-INF/container.xml b/...l-identifier-error/META-INF/container.xml → ...r-disallowed-error/META-INF/container.xml
diff --git a/src/test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/mimetype b/src/test/resources/epub3/files/epub/xml-external-identifier-disallowed-error/mimetype
@@ -0,0 +1 @@
+application/epub+zip
diff --git a/src/test/resources/epub3/resources-publication.feature b/src/test/resources/epub3/resources-publication.feature
@@ -254,8 +254,17 @@ Feature: EPUB 3 ▸ Publication Resources ▸ Full Publication Checks
 
   ##  3.3 XML Conformance
 
-  Scenario: Report an NCX file with a DOCTYPE declaration including the external identifier (issue 305)
-    When checking EPUB 'xml-ncx-doctype-external-identifier-error'
+  Scenario: Verify DOCTYPE declarations with allowed external identifiers
+    When checking EPUB 'xml-external-identifier-allowed-valid'
+    Then no errors or warnings are reported
+
+  Scenario: Report a DOCTYPE declaration with an allowed external identifier but not on the expected media type
+    When checking EPUB 'xml-external-identifier-bad-mediatype-error'
+    Then error OPF-073 is reported
+    And no other errors or warnings are reported
+
+  Scenario: Report a DOCTYPE declaration with an external identifier that is not allowed
+    When checking EPUB 'xml-external-identifier-disallowed-error'
     Then error OPF-073 is reported
     And no other errors or warnings are reported