[cpd] Add ignore literals and identifiers capability to C++

pmd · May 24, 2024 · 47d5fbf · 47d5fbf
1 parent 2c64541
commit 47d5fbf
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 4 deletions.
diff --git a/docs/pages/pmd/userdocs/cpd/cpd.md b/docs/pages/pmd/userdocs/cpd/cpd.md
@@ -158,7 +158,7 @@ exactly identical.
     {% include custom/cli_option_row.html options="--ignore-literals"
                description="Ignore literal values such as numbers and strings when comparing text.
                             By default, literals are not ignored."
-               languages="Java"
+               languages="Java, C++"
     %}
     {% include custom/cli_option_row.html options="--ignore-literal-sequences"
                description="Ignore sequences of literals such as list initializers.
@@ -168,7 +168,7 @@ exactly identical.
     {% include custom/cli_option_row.html options="--ignore-identifiers"
                description="Ignore names of classes, methods, variables, constants, etc. when comparing text.
                             By default, identifier names are not ignored."
-               languages="Java"
+               languages="Java, C++"
     %}
     {% include custom/cli_option_row.html options="--ignore-annotations"
                description="Ignore language annotations (Java) or attributes (C#) when comparing text.

diff --git a/pmd-cpp/src/main/java/net/sourceforge/pmd/lang/cpp/CppLanguageModule.java b/pmd-cpp/src/main/java/net/sourceforge/pmd/lang/cpp/CppLanguageModule.java
@@ -47,6 +47,8 @@ public LanguagePropertyBundle newPropertyBundle() {
         LanguagePropertyBundle bundle = super.newPropertyBundle();
         bundle.definePropertyDescriptor(CpdLanguageProperties.CPD_IGNORE_LITERAL_SEQUENCES);
         bundle.definePropertyDescriptor(CpdLanguageProperties.CPD_IGNORE_LITERAL_AND_IDENTIFIER_SEQUENCES);
+        bundle.definePropertyDescriptor(CpdLanguageProperties.CPD_ANONYMIZE_IDENTIFIERS);
+        bundle.definePropertyDescriptor(CpdLanguageProperties.CPD_ANONYMIZE_LITERALS);
         bundle.definePropertyDescriptor(CPD_SKIP_BLOCKS);
         return bundle;
     }

diff --git a/pmd-cpp/src/main/java/net/sourceforge/pmd/lang/cpp/cpd/CppCpdLexer.java b/pmd-cpp/src/main/java/net/sourceforge/pmd/lang/cpp/cpd/CppCpdLexer.java
@@ -9,8 +9,9 @@
 import org.apache.commons.lang3.StringUtils;
 
 import net.sourceforge.pmd.cpd.CpdLanguageProperties;
-import net.sourceforge.pmd.cpd.impl.CpdLexerBase;
+import net.sourceforge.pmd.cpd.TokenFactory;
 import net.sourceforge.pmd.cpd.impl.JavaCCTokenFilter;
+import net.sourceforge.pmd.cpd.impl.JavaccCpdLexer;
 import net.sourceforge.pmd.lang.LanguagePropertyBundle;
 import net.sourceforge.pmd.lang.TokenManager;
 import net.sourceforge.pmd.lang.ast.impl.javacc.CharStream;
@@ -26,17 +27,21 @@
  *
  * <p>Note: This class has been called CPPTokenizer in PMD 6</p>.
  */
-public class CppCpdLexer extends CpdLexerBase<JavaccToken> {
+public class CppCpdLexer extends JavaccCpdLexer {
 
     private boolean skipBlocks;
     private Pattern skipBlocksStart;
     private Pattern skipBlocksEnd;
     private final boolean ignoreIdentifierAndLiteralSeqences;
     private final boolean ignoreLiteralSequences;
+    private final boolean ignoreLiterals;
+    private final boolean ignoreIdentifiers;
 
     public CppCpdLexer(LanguagePropertyBundle cppProperties) {
         ignoreLiteralSequences = cppProperties.getProperty(CpdLanguageProperties.CPD_IGNORE_LITERAL_SEQUENCES);
         ignoreIdentifierAndLiteralSeqences = cppProperties.getProperty(CpdLanguageProperties.CPD_IGNORE_LITERAL_AND_IDENTIFIER_SEQUENCES);
+        ignoreLiterals = cppProperties.getProperty(CpdLanguageProperties.CPD_ANONYMIZE_LITERALS);
+        ignoreIdentifiers = cppProperties.getProperty(CpdLanguageProperties.CPD_ANONYMIZE_IDENTIFIERS);
         String skipBlocksPattern = cppProperties.getProperty(CppLanguageModule.CPD_SKIP_BLOCKS);
         if (StringUtils.isNotBlank(skipBlocksPattern)) {
             skipBlocks = true;
@@ -73,6 +78,23 @@ protected TokenManager<JavaccToken> filterTokenStream(final TokenManager<JavaccT
         return new CppTokenFilter(tokenManager, ignoreLiteralSequences, ignoreIdentifierAndLiteralSeqences);
     }
 
+    @Override
+    protected void processToken(TokenFactory tokenEntries, JavaccToken currentToken) {
+        int kind = currentToken.getKind();
+        String image = currentToken.getImage();
+
+        boolean isLiteral = kind == CppTokenKinds.STRING || kind == CppTokenKinds.CHARACTER || kind == CppTokenKinds.DECIMAL_INT_LITERAL || kind == CppTokenKinds.HEXADECIMAL_INT_LITERAL || kind == CppTokenKinds.OCTAL_INT_LITERAL || kind == CppTokenKinds.FLOAT_LITERAL || kind == CppTokenKinds.BINARY_INT_LITERAL || kind == CppTokenKinds.ZERO;
+        if (ignoreLiterals && isLiteral) {
+            image = CppTokenKinds.describe(kind);
+        }
+
+        if (ignoreIdentifiers && (kind == CppTokenKinds.ID)) {
+            image = CppTokenKinds.describe(kind);
+        }
+
+        tokenEntries.recordToken(image, currentToken.getReportLocation());
+    }
+
     private static class CppTokenFilter extends JavaCCTokenFilter {
 
         private final boolean ignoreLiteralSequences;