Add support for Unicode for JSP/JSF.

Add support to not parse body of <script> for JSP/JSF.
Add new NoInlineScript rule for JSP/JSF.

git-svn-id: https://pmd.svn.sourceforge.net/svnroot/pmd/branches/pmd/4.2.x@7015 51baf565-9d33-0410-a72c-fc3788e3496d

pmd/bin/build.xml

@@ -392,8 +392,16 @@
 		<delete file="src/net/sourceforge/pmd/jsp/ast/ParseException.java" />
 		<delete file="src/net/sourceforge/pmd/jsp/ast/CharStream.java" />
 		<jjtree target="etc/grammar/JspParser.jjt" outputdirectory="src/net/sourceforge/pmd/jsp/ast/" javacchome="${javacc-home.path}" />
+
+		<!-- Generate Unicode w/o Unicode Escapes CharStream implementation -->
+		<javacc static="false" usercharstream="false" unicodeinput="true" javaunicodeescape="false" target="src/net/sourceforge/pmd/jsp/ast/JspParser.jj" outputdirectory="src/net/sourceforge/pmd/jsp/ast" javacchome="${javacc-home.path}" />
+		<replace file="src/net/sourceforge/pmd/jsp/ast/SimpleCharStream.java" token="public class SimpleCharStream" value="public class JspCharStream implements CharStream" />
+		<replace file="src/net/sourceforge/pmd/jsp/ast/SimpleCharStream.java" token="public SimpleCharStream" value="public JspCharStream" />
+		<move file="src/net/sourceforge/pmd/jsp/ast/SimpleCharStream.java" tofile="src/net/sourceforge/pmd/jsp/ast/JspCharStream.java"/>
+		<delete file="src/net/sourceforge/pmd/jsp/ast/JspParser.java" />
+
 		<!-- Ensure generated using CharStream interface -->
-		<javacc static="false" usercharstream="true" target="src/net/sourceforge/pmd/jsp/ast/JspParser.jj" outputdirectory="src/net/sourceforge/pmd/jsp/ast" javacchome="${javacc-home.path}" />
+		<javacc static="false" usercharstream="true" unicodeinput="true" javaunicodeescape="false" target="src/net/sourceforge/pmd/jsp/ast/JspParser.jj" outputdirectory="src/net/sourceforge/pmd/jsp/ast" javacchome="${javacc-home.path}" />
 
 		<delete file="src/net/sourceforge/pmd/jsp/ast/JspParser.jj" />
 		<replace file="src/net/sourceforge/pmd/jsp/ast/JspParserTokenManager.java" token="throw new Error" value="throw new RuntimeException" />

pmd/etc/changelog.txt

@@ -13,6 +13,8 @@ Fixed bug 2904832 - Type resolution not working for ASTType when using an inner
 
 Correct -benchmark reporting of Rule visits via the RuleChain
 Fix issue with Type Resolution incorrectly handling of Classes with same name as a java.lang Class.
+The JSP/JSF parser can now parse Unicode input.
+The JSP/JSP parser can now handle <script>...</script> tags.  The AST HtmlScript node contains the content.
 
 Dependencies updates: asm updated to 3.2
 
@@ -23,6 +25,7 @@ New rule:
 	Controversial : AvoidLiteralsInIfCondition (patch 2591627), UseConcurrentHashMap
 	StrictExceptions : AvoidCatchingGenericException, AvoidLosingExceptionInformation
 	Naming : GenericsNaming
+	JSP: NoInlineScript
 
 February 08, 2009 - 4.2.5:
 

pmd/etc/grammar/JspParser.jjt

@@ -1,13 +1,15 @@
-    /* JSP Parser for PMD.
+/* JSP Parser for PMD.
  * It supports supports more-or-less well written JSP files.
  * The JSP Document style is supported, except for inline DTD.
  * The JSP Page style (<% ... %>) is supported.
  * Java code is not parsed.
+ * Script code inside <script> ... </script> is not parsed.
  */
 
 options { 
 	USER_CHAR_STREAM = true;
 	NODE_USES_PARSER=true;
+	UNICODE_INPUT=true;
 	  
 	IGNORE_CASE = true; 
 	STATIC = false; 
@@ -19,7 +21,7 @@ options {
 PARSER_BEGIN(JspParser)
 /** 
  * JSP Parser for PMD.
- * @author Pieter – Application Engineers NV/SA – http://www.ae.be
+ * @author Pieter – Application Engineers NV/SA – http://www.ae.be
  */
 
 package net.sourceforge.pmd.jsp.ast;
@@ -78,6 +80,7 @@ PARSER_END(JspParser)
  * - JspExpressionState : inside an expression <%= ... %>
  * - JspDeclarationState : inside a declaration <%! ... %>
  * - JspCommentState : inside a comment <%-- ... --%>
+ * - HtmlScriptContentState : inside an HTML script <script> ... </script>
  */
 
 
@@ -263,6 +266,18 @@ PARSER_END(JspParser)
 | < COMMENT_TEXT: (~[]) >
 }
 
+<HtmlScriptContentState> TOKEN :
+{
+	<HTML_SCRIPT_CONTENT: (~[]) >
+	| <HTML_SCRIPT_END_TAG : "</script" | "</Script" | "</SCRIPT">
+	{
+		// We've done a custom lookahead for the closing </script> and found it.
+		// Put it back into the input stream, so it can be processed normally.
+		input_stream.backup(8);
+		image.setLength(image.length() - 8); // kill the "<script"
+		matchedToken.image = image.toString();
+	} : AfterTagState
+}
 
 /** ******************************************************************** */
 /** *************************  JSP GRAMMAR  **************************** */
@@ -498,9 +513,20 @@ void Element() :
     (Attribute())*
     ( 
     	(<TAG_END> 
-    		{ jjtThis.setEmpty(false); }
+    		{
+    			jjtThis.setEmpty(false);
+
+		    	// Content in a <script> element needs special treatment (like a comment or CDataSection).
+		    	// Tell the TokenManager to start looking for the body of a script element.  In this
+		    	// state all text will be consumed by the next token up to the closing </script> tag.
+		    	// This is a context sensitive switch for the token manager, not something one can
+		    	// express using normal JavaCC syntax.  Hence the hoop jumping.
+				if ("script".equalsIgnoreCase(startTagName.image)) {
+					token_source.SwitchTo(HtmlScriptContentState);
+				}
+    		}
     	
-    	(Content())? 
+   		(HtmlScript() | Content())?
     	
     	<ENDTAG_START> 
     	endTagName = <TAG_NAME> 
@@ -645,3 +671,16 @@ void DoctypeExternalId() :
 				{ jjtThis.setUri(quoteContent(systemLiteral.image)); }
 		)
 }
+
+void HtmlScript() :
+{
+	StringBuffer content = new StringBuffer();
+	Token t;
+}
+{
+	(t = <HTML_SCRIPT_CONTENT>  { content.append(t.image); })*
+	<HTML_SCRIPT_END_TAG>
+		{
+			jjtThis.setImage(content.toString().trim());
+		}
+}

pmd/regress/test/net/sourceforge/pmd/jsp/ast/JspDocStyleTest.java

@@ -1,6 +1,13 @@
 package test.net.sourceforge.pmd.jsp.ast;
 
 import static org.junit.Assert.assertEquals;
+
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.List;
+import java.util.Set;
+
 import net.sourceforge.pmd.jsp.ast.ASTAttribute;
 import net.sourceforge.pmd.jsp.ast.ASTAttributeValue;
 import net.sourceforge.pmd.jsp.ast.ASTCData;
@@ -8,14 +15,9 @@ import net.sourceforge.pmd.jsp.ast.ASTCommentTag;
 import net.sourceforge.pmd.jsp.ast.ASTDoctypeDeclaration;
 import net.sourceforge.pmd.jsp.ast.ASTDoctypeExternalId;
 import net.sourceforge.pmd.jsp.ast.ASTElement;
+import net.sourceforge.pmd.jsp.ast.ASTHtmlScript;
 
 import org.junit.Test;
-
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.Comparator;
-import java.util.List;
-import java.util.Set;
 /**
  * Test parsing of a JSP in document style, by checking the generated AST.
  * 
@@ -156,6 +158,17 @@ public class JspDocStyleTest extends AbstractJspNodesTst {
 		ASTCommentTag comment = (ASTCommentTag) comments.iterator().next();
 		assertEquals("Correct comment content expected!", "comment", comment.getImage());
 	}
+    
+    /**
+     * Test parsing of HTML <script> element.
+     */
+    @Test
+    public void testHtmlScript() {
+	Set scripts = getNodes(ASTHtmlScript.class, TEST_HTML_SCRIPT);
+	assertEquals("One script expected!", 1, scripts.size());
+	ASTHtmlScript script = (ASTHtmlScript) scripts.iterator().next();
+	assertEquals("Correct script content expected!", "Script!", script.getImage());
+    }
 
 	private static final String TEST_SIMPLEST_HTML = "<html/>";
 
@@ -173,6 +186,9 @@ public class JspDocStyleTest extends AbstractJspNodesTst {
 	private static final String TEST_ATTRIBUTE_VALUE_CONTAINING_HASH = 
 		"<tag:if something=\"#yes#\" foo=\"CREATE\">  <a href=\"#\">foo</a> </tag:if>";
 
+	private static final String TEST_HTML_SCRIPT =
+		"<html><head><script>Script!</script></head></html>";
+
     public static junit.framework.Test suite() {
         return new junit.framework.JUnit4TestAdapter(JspDocStyleTest.class);
     }

pmd/regress/test/net/sourceforge/pmd/jsp/rules/NoInlineScriptTest.java

@@ -0,0 +1,17 @@
+package test.net.sourceforge.pmd.jsp.rules;
+
+import org.junit.Before;
+
+import test.net.sourceforge.pmd.testframework.SimpleAggregatorTst;
+
+public class NoInlineScriptTest extends SimpleAggregatorTst {
+
+    @Before
+    public void setUp() {
+        addRule("jsp", "NoInlineScript");
+    }
+
+    public static junit.framework.Test suite() {
+        return new junit.framework.JUnit4TestAdapter(NoInlineScriptTest.class);
+    }
+}

pmd/regress/test/net/sourceforge/pmd/jsp/rules/xml/NoInlineScript.xml

@@ -0,0 +1,53 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<test-data>
+    <test-code>
+        <description><![CDATA[
+A violation.
+     ]]></description>
+        <expected-problems>1</expected-problems>
+        <code><![CDATA[
+<html><head><script>Some script</script></head></html>
+     ]]></code>
+        <source-type>jsp</source-type>
+    </test-code>
+    <test-code>
+        <description><![CDATA[
+A violation.
+     ]]></description>
+        <expected-problems>1</expected-problems>
+        <code><![CDATA[
+<html><head><script src="script location">Some script</script></head></html>
+     ]]></code>
+        <source-type>jsp</source-type>
+    </test-code>
+    <test-code>
+        <description><![CDATA[
+Ok, no 'src' with empty body.
+     ]]></description>
+        <expected-problems>0</expected-problems>
+        <code><![CDATA[
+<html><head><script></script></head></html>
+     ]]></code>
+        <source-type>jsp</source-type>
+    </test-code>
+    <test-code>
+        <description><![CDATA[
+Ok, only 'src' with empty whitespace body.
+     ]]></description>
+        <expected-problems>0</expected-problems>
+        <code><![CDATA[
+<html><head><script src="script location"> </script></head></html>
+     ]]></code>
+        <source-type>jsp</source-type>
+    </test-code>
+    <test-code>
+        <description><![CDATA[
+Ok, only 'src' without body.
+     ]]></description>
+        <expected-problems>0</expected-problems>
+        <code><![CDATA[
+<html><head><script src="script location"/></head></html>
+     ]]></code>
+        <source-type>jsp</source-type>
+    </test-code>
+</test-data>

pmd/rulesets/basic-jsp.xml

@@ -261,7 +261,39 @@ and
                 
                 <%@ page contentType="text/html; charset=UTF-8" pageEncoding="UTF-8" %>
                     
-                <meta http-equiv="Content-Type"  content="text/html; charset=UTF-8" />
+                <meta http-equiv="Content-Type"  content="text/html; charset=UTF-8" />
+            ]]>
+        </example>
+    </rule>
+    
+    <rule
+        name="NoInlineScript"
+        class="net.sourceforge.pmd.rules.XPathRule"
+        message="Avoiding inlining HTML script content"
+        externalInfoUrl="http://pmd.sourceforge.net/rules/basic-jsp.html#JspEncoding">
+        <description>
+            <![CDATA[
+Avoid inlining HTML script content.  Consider externalizing the HTML script using the 'src' attribute on the <script> element.
+Externalized script could be reused between pages.  Browsers can also cache the script, reducing overall download bandwidth.
+            ]]>
+        </description>
+        <priority>3</priority>
+        <properties>
+            <property name="xpath">
+                <value>
+                    <![CDATA[
+//HtmlScript[@Image != '']
+                    ]]>
+                </value>
+            </property>
+        </properties>
+        <example>
+            <![CDATA[
+                Most browsers should be able to interpret the following headers:
+                
+                <%@ page contentType="text/html; charset=UTF-8" pageEncoding="UTF-8" %>
+                    
+                <meta http-equiv="Content-Type"  content="text/html; charset=UTF-8" />
             ]]>
         </example>
     </rule>

pmd/src/net/sourceforge/pmd/jsp/ast/ASTHtmlScript.java

@@ -0,0 +1,20 @@
+/* Generated By:JJTree: Do not edit this line. ASTHtmlScript.java Version 4.1 */
+/* JavaCCOptions:MULTI=true,NODE_USES_PARSER=true,VISITOR=true,TRACK_TOKENS=false,NODE_PREFIX=AST,NODE_EXTENDS=,NODE_FACTORY= */
+package net.sourceforge.pmd.jsp.ast;
+
+public class ASTHtmlScript extends SimpleNode {
+  public ASTHtmlScript(int id) {
+    super(id);
+  }
+
+  public ASTHtmlScript(JspParser p, int id) {
+    super(p, id);
+  }
+
+
+  /** Accept the visitor. **/
+  public Object jjtAccept(JspParserVisitor visitor, Object data) {
+    return visitor.visit(this, data);
+  }
+}
+/* JavaCC - OriginalChecksum=8170f25fd45a522c3d56439f6d24e048 (do not edit this line) */

pmd/src/net/sourceforge/pmd/jsp/ast/JspParserConstants.java

@@ -1,7 +1,7 @@
 /* Generated By:JJTree&JavaCC: Do not edit this line. JspParserConstants.java */
 /** 
  * JSP Parser for PMD.
- * @author Pieter � Application Engineers NV/SA � http://www.ae.be
+ * @author Pieter – Application Engineers NV/SA – http://www.ae.be
  */
 
 package net.sourceforge.pmd.jsp.ast;
@@ -163,41 +163,47 @@ public interface JspParserConstants {
   int COMMENT_END = 75;
   /** RegularExpression Id. */
   int COMMENT_TEXT = 76;
+  /** RegularExpression Id. */
+  int HTML_SCRIPT_CONTENT = 77;
+  /** RegularExpression Id. */
+  int HTML_SCRIPT_END_TAG = 78;
 
   /** Lexical state. */
-  int CommentState = 0;
+  int HtmlScriptContentState = 0;
   /** Lexical state. */
-  int AttrValueBetweenDoubleQuotesState = 1;
+  int CommentState = 1;
   /** Lexical state. */
-  int AttrValueBetweenSingleQuotesState = 2;
+  int AttrValueBetweenDoubleQuotesState = 2;
   /** Lexical state. */
-  int StartTagState = 3;
+  int AttrValueBetweenSingleQuotesState = 3;
   /** Lexical state. */
-  int CDataState = 4;
+  int StartTagState = 4;
   /** Lexical state. */
-  int DocTypeExternalIdState = 5;
+  int CDataState = 5;
   /** Lexical state. */
-  int DocTypeState = 6;
+  int DocTypeExternalIdState = 6;
   /** Lexical state. */
-  int JspCommentState = 7;
+  int DocTypeState = 7;
   /** Lexical state. */
-  int JspDeclarationState = 8;
+  int JspCommentState = 8;
   /** Lexical state. */
-  int JspExpressionState = 9;
+  int JspDeclarationState = 9;
   /** Lexical state. */
-  int JspScriptletState = 10;
+  int JspExpressionState = 10;
   /** Lexical state. */
-  int InTagState = 11;
+  int JspScriptletState = 11;
   /** Lexical state. */
-  int AfterTagState = 12;
+  int InTagState = 12;
   /** Lexical state. */
-  int AttrValueState = 13;
+  int AfterTagState = 13;
   /** Lexical state. */
-  int JspDirectiveAttributesState = 14;
+  int AttrValueState = 14;
   /** Lexical state. */
-  int JspDirectiveState = 15;
+  int JspDirectiveAttributesState = 15;
   /** Lexical state. */
-  int DEFAULT = 16;
+  int JspDirectiveState = 16;
+  /** Lexical state. */
+  int DEFAULT = 17;
 
   /** Literal token values. */
   String[] tokenImage = {
@@ -278,6 +284,8 @@ public interface JspParserConstants {
     "<DOLLAR_OR_HASH_DOUBLE_QUOTE>",
     "<COMMENT_END>",
     "<COMMENT_TEXT>",
+    "<HTML_SCRIPT_CONTENT>",
+    "<HTML_SCRIPT_END_TAG>",
   };
 
 }

pmd/src/net/sourceforge/pmd/jsp/ast/JspParserTreeConstants.java

@@ -25,6 +25,7 @@ public interface JspParserTreeConstants
   public int JJTDECLARATION = 19;
   public int JJTDOCTYPEDECLARATION = 20;
   public int JJTDOCTYPEEXTERNALID = 21;
+  public int JJTHTMLSCRIPT = 22;
 
 
   public String[] jjtNodeName = {
@@ -50,6 +51,7 @@ public interface JspParserTreeConstants
     "Declaration",
     "DoctypeDeclaration",
     "DoctypeExternalId",
+    "HtmlScript",
   };
 }
-/* JavaCC - OriginalChecksum=d7bde07aef0db02727c15b118c4c677f (do not edit this line) */
+/* JavaCC - OriginalChecksum=e88447ad3f3129b84d706554cb6c3195 (do not edit this line) */

pmd/src/net/sourceforge/pmd/jsp/ast/JspParserVisitor.java

@@ -25,5 +25,6 @@ public interface JspParserVisitor
   public Object visit(ASTDeclaration node, Object data);
   public Object visit(ASTDoctypeDeclaration node, Object data);
   public Object visit(ASTDoctypeExternalId node, Object data);
+  public Object visit(ASTHtmlScript node, Object data);
 }
-/* JavaCC - OriginalChecksum=85c3652943205ef4624387751f8ca0e6 (do not edit this line) */
+/* JavaCC - OriginalChecksum=04069d9511f38dbdc673d112c08dd375 (do not edit this line) */

pmd/src/net/sourceforge/pmd/jsp/ast/JspParserVisitorAdapter.java

@@ -92,4 +92,7 @@ public class JspParserVisitorAdapter implements JspParserVisitor {
         return visit((SimpleNode) node, data);
     }
 
+    public Object visit(ASTHtmlScript node, Object data) {
+        return visit((SimpleNode) node, data);
+    }
 }

pmd/src/net/sourceforge/pmd/jsp/rules/AbstractJspRule.java

@@ -22,6 +22,7 @@ import net.sourceforge.pmd.jsp.ast.ASTDoctypeDeclaration;
 import net.sourceforge.pmd.jsp.ast.ASTDoctypeExternalId;
 import net.sourceforge.pmd.jsp.ast.ASTElExpression;
 import net.sourceforge.pmd.jsp.ast.ASTElement;
+import net.sourceforge.pmd.jsp.ast.ASTHtmlScript;
 import net.sourceforge.pmd.jsp.ast.ASTJspComment;
 import net.sourceforge.pmd.jsp.ast.ASTJspDeclaration;
 import net.sourceforge.pmd.jsp.ast.ASTJspDirective;
@@ -45,7 +46,7 @@ public abstract class AbstractJspRule extends CommonAbstractRule implements
 
 	/**
 	 * Adds a violation to the report.
-	 * 
+	 *
 	 * @param data
 	 *            the RuleContext
 	 * @param node
@@ -58,7 +59,7 @@ public abstract class AbstractJspRule extends CommonAbstractRule implements
 
 	/**
 	 * Adds a violation to the report.
-	 * 
+	 *
 	 * @param data
 	 *            the RuleContext
 	 * @param node
@@ -75,7 +76,7 @@ public abstract class AbstractJspRule extends CommonAbstractRule implements
 
 	/**
 	 * Adds a violation to the report.
-	 * 
+	 *
 	 * @param data
 	 *            the RuleContext
 	 * @param node
@@ -92,7 +93,7 @@ public abstract class AbstractJspRule extends CommonAbstractRule implements
 
 	/**
 	 * Adds a violation to the report.
-	 * 
+	 *
 	 * @param data
 	 *            the RuleContext
 	 * @param node
@@ -213,4 +214,8 @@ public abstract class AbstractJspRule extends CommonAbstractRule implements
 	public Object visit(ASTDoctypeExternalId node, Object data) {
 		return visit((SimpleNode)node, data);
 	}
+
+	public Object visit(ASTHtmlScript node, Object data) {
+		return visit((SimpleNode)node, data);
+	}
 }