feat: allow informative schema checking (reported as new RSC-024 and RSC-025)

rdeltour · rdeltour · commit a2516f060904 · 2021-11-14T15:48:51.000+01:00
An `XMLValidator` instance can now be instantiated with a boolean
flag to tell if the unerlying schema is normative or not.

If a schema is not normative, the `XMLParser` uses an informative-only `ErrorHandler`,
that reports warnings as "RSC-024" (USAGE), and errors as "RSC-025" (USAGE).
Fatal errors are still reported as "RSC-016" (FATAL).
diff --git a/src/main/java/com/adobe/epubcheck/messages/DefaultSeverities.java b/src/main/java/com/adobe/epubcheck/messages/DefaultSeverities.java
@@ -325,6 +325,8 @@ private void initialize()
     severities.put(MessageId.RSC_021, Severity.ERROR);
     severities.put(MessageId.RSC_022, Severity.INFO);
     severities.put(MessageId.RSC_023, Severity.WARNING);
+    severities.put(MessageId.RSC_024, Severity.USAGE);
+    severities.put(MessageId.RSC_025, Severity.USAGE);
 
     // Scripting
     severities.put(MessageId.SCP_001, Severity.USAGE);
diff --git a/src/main/java/com/adobe/epubcheck/messages/MessageId.java b/src/main/java/com/adobe/epubcheck/messages/MessageId.java
@@ -319,6 +319,8 @@ public enum MessageId implements Comparable<MessageId>
   RSC_021("RSC-021"),
   RSC_022("RSC-022"),
   RSC_023("RSC-023"),
+  RSC_024("RSC-024"),
+  RSC_025("RSC-025"),
 
   // Messages relating to scripting
   SCP_001("SCP-001"),
diff --git a/src/main/java/com/adobe/epubcheck/xml/XMLParser.java b/src/main/java/com/adobe/epubcheck/xml/XMLParser.java
@@ -41,6 +41,7 @@
 import org.xml.sax.Attributes;
 import org.xml.sax.ContentHandler;
 import org.xml.sax.DTDHandler;
+import org.xml.sax.ErrorHandler;
 import org.xml.sax.InputSource;
 import org.xml.sax.Locator;
 import org.xml.sax.SAXException;
@@ -68,6 +69,46 @@
 
 public class XMLParser extends DefaultHandler implements LexicalHandler, DeclHandler
 {
+
+  private static final class UsageErrorHandler implements ErrorHandler
+  {
+    private final ValidationContext context;
+
+    public UsageErrorHandler(ValidationContext context)
+    {
+      this.context = context;
+    }
+
+    @Override
+    public void warning(SAXParseException exception)
+      throws SAXException
+    {
+      context.report.message(MessageId.RSC_024,
+          EPUBLocation.create(context.path, exception.getLineNumber(), exception.getColumnNumber()),
+          exception.getMessage());
+    }
+
+    @Override
+    public void error(SAXParseException exception)
+      throws SAXException
+    {
+      context.report.message(MessageId.RSC_025,
+          EPUBLocation.create(context.path, exception.getLineNumber(), exception.getColumnNumber()),
+          exception.getMessage());
+    }
+
+    @Override
+    public void fatalError(SAXParseException exception)
+      throws SAXException
+    {
+      context.report.message(MessageId.RSC_016,
+          EPUBLocation.create(context.path, exception.getLineNumber(), exception.getColumnNumber()),
+          exception.getMessage());
+
+    }
+
+  }
+
   private static final String SAXPROP_LEXICAL_HANDLER = "https://xml.org/sax/properties/lexical-handler";
   private static final String SAXPROP_DECL_HANDLER = "https://xml.org/sax/properties/declaration-handler";
   private SAXParser parser;
@@ -158,8 +199,9 @@ public void addXMLHandler(XMLHandler handler)
   public void addValidator(XMLValidator xv)
   {
     PropertyMapBuilder propertyMapBuilder = new PropertyMapBuilder();
-    propertyMapBuilder.put(ValidateProperty.ERROR_HANDLER, this);
-    Validator validator = xv.schema.createValidator(propertyMapBuilder.toPropertyMap());
+    ErrorHandler eh = xv.isNormative() ? this : new UsageErrorHandler(context);
+    propertyMapBuilder.put(ValidateProperty.ERROR_HANDLER, eh);
+    Validator validator = xv.getSchema().createValidator(propertyMapBuilder.toPropertyMap());
     ContentHandler contentHandler = validator.getContentHandler();
     if (contentHandler != null)
     {
diff --git a/src/main/java/com/adobe/epubcheck/xml/XMLValidator.java b/src/main/java/com/adobe/epubcheck/xml/XMLValidator.java
@@ -28,7 +28,6 @@
 import java.net.URISyntaxException;
 import java.net.URL;
 
-import javax.xml.transform.TransformerException;
 import javax.xml.transform.TransformerFactory;
 
 import org.idpf.epubcheck.util.saxon.ColumnNumberFunction;
@@ -55,18 +54,16 @@
 
 import net.sf.saxon.Configuration;
 import net.sf.saxon.TransformerFactoryImpl;
-import net.sf.saxon.lib.FeatureKeys;
-import net.sf.saxon.lib.StandardErrorListener;
 import net.sf.saxon.sxpath.IndependentContext;
 import net.sf.saxon.sxpath.XPathStaticContext;
 import net.sf.saxon.trans.SymbolicName;
-import net.sf.saxon.trans.XPathException;
 
 
 public class XMLValidator
 {
 
-  Schema schema;
+  private final Schema schema;
+  private final boolean isNormative;
 
   /**
    * Basic Resolver from Jing modified to add support for resolving zip and
@@ -223,9 +220,10 @@ public void warning(SAXParseException exception) throws
     }
 
   }
-
-  public XMLValidator(String schemaName)
+  
+  public XMLValidator(String schemaName, boolean isNormative)
   {
+    this.isNormative = isNormative;
     try
     {
       String resourcePath = ResourceUtil.getResourcePath(schemaName);
@@ -271,4 +269,12 @@ public XMLValidator(String schemaName)
       throw new Error("Internal error: " + e + " " + schemaName);
     }
   }
+  
+  public Schema getSchema() {
+    return schema;
+  }
+  
+  public boolean isNormative() {
+    return isNormative;
+  }
 }
diff --git a/src/main/java/com/adobe/epubcheck/xml/XMLValidators.java b/src/main/java/com/adobe/epubcheck/xml/XMLValidators.java
@@ -55,7 +55,12 @@ public enum XMLValidators
 
   private XMLValidators(String schemaName)
   {
-    val = new XMLValidator(schemaName);
+    this(schemaName, true);
+  }
+
+  private XMLValidators(String schemaName, boolean isNormative)
+  {
+    this.val = new XMLValidator(schemaName, isNormative);
   }
 
   public XMLValidator get()
diff --git a/src/main/resources/com/adobe/epubcheck/messages/MessageBundle.properties b/src/main/resources/com/adobe/epubcheck/messages/MessageBundle.properties
@@ -337,6 +337,8 @@ RSC_020="%1$s" is not a valid URI.
 RSC_021=A Search Key Map Document must point to Content Documents ("%1$s" was not found in the spine).
 RSC_022=Cannot check image details (requires Java version 7 or higher).
 RSC_023=Couldn’t parse host of URL "%1$s" (probably due to disallowed characters or missing slashes after the protocol)
+RSC_024=Informative parsing warning: %1$s
+RSC_025=Informative parsing error: %1$s
 
 #Scripting
 SCP_001=Use of Javascript eval() function in EPUB scripts is a security risk.

Original file line number	Diff line number	Diff line change
`@@ -28,7 +28,6 @@`
`28`	`28`	`import java.net.URISyntaxException;`
`29`	`29`	`import java.net.URL;`
`30`	`30`
`31`		`-import javax.xml.transform.TransformerException;`
`32`	`31`	`import javax.xml.transform.TransformerFactory;`
`33`	`32`
`34`	`33`	`import org.idpf.epubcheck.util.saxon.ColumnNumberFunction;`
`@@ -55,18 +54,16 @@`
`55`	`54`
`56`	`55`	`import net.sf.saxon.Configuration;`
`57`	`56`	`import net.sf.saxon.TransformerFactoryImpl;`
`58`		`-import net.sf.saxon.lib.FeatureKeys;`
`59`		`-import net.sf.saxon.lib.StandardErrorListener;`
`60`	`57`	`import net.sf.saxon.sxpath.IndependentContext;`
`61`	`58`	`import net.sf.saxon.sxpath.XPathStaticContext;`
`62`	`59`	`import net.sf.saxon.trans.SymbolicName;`
`63`		`-import net.sf.saxon.trans.XPathException;`
`64`	`60`
`65`	`61`
`66`	`62`	`public class XMLValidator`
`67`	`63`	`{`
`68`	`64`
`69`		`- Schema schema;`
	`65`	`+ private final Schema schema;`
	`66`	`+ private final boolean isNormative;`
`70`	`67`
`71`	`68`	`/**`
`72`	`69`	`* Basic Resolver from Jing modified to add support for resolving zip and`
`@@ -223,9 +220,10 @@ public void warning(SAXParseException exception) throws`
`223`	`220`	`}`
`224`	`221`
`225`	`222`	`}`
`226`		`-`
`227`		`- public XMLValidator(String schemaName)`
	`223`	`+`
	`224`	`+ public XMLValidator(String schemaName, boolean isNormative)`
`228`	`225`	`{`
	`226`	`+ this.isNormative = isNormative;`
`229`	`227`	`try`
`230`	`228`	`{`
`231`	`229`	`String resourcePath = ResourceUtil.getResourcePath(schemaName);`
`@@ -271,4 +269,12 @@ public XMLValidator(String schemaName)`
`271`	`269`	`throw new Error("Internal error: " + e + " " + schemaName);`
`272`	`270`	`}`
`273`	`271`	`}`
	`272`	`+`
	`273`	`+ public Schema getSchema() {`
	`274`	`+ return schema;`
	`275`	`+ }`
	`276`	`+`
	`277`	`+ public boolean isNormative() {`
	`278`	`+ return isNormative;`
	`279`	`+ }`
`274`	`280`	`}`
Original file line number	Diff line number	Diff line change
`@@ -55,7 +55,12 @@ public enum XMLValidators`
`55`	`55`
`56`	`56`	`private XMLValidators(String schemaName)`
`57`	`57`	`{`
`58`		`- val = new XMLValidator(schemaName);`
	`58`	`+ this(schemaName, true);`
	`59`	`+ }`
	`60`	`+`
	`61`	`+ private XMLValidators(String schemaName, boolean isNormative)`
	`62`	`+ {`
	`63`	`+ this.val = new XMLValidator(schemaName, isNormative);`
`59`	`64`	`}`
`60`	`65`
`61`	`66`	`public XMLValidator get()`