Fix for #1060: add error marker for control and format characters

groovy · Mar 16, 2020 · c2425bd · c2425bd
1 parent 69f3f86
commit c2425bd
Show file tree

Hide file tree

Showing 17 changed files with 7,962 additions and 7,803 deletions.
diff --git a/...vy.core.tests.compiler/src/org/eclipse/jdt/groovy/core/tests/basic/GroovySimpleTests.java b/...vy.core.tests.compiler/src/org/eclipse/jdt/groovy/core/tests/basic/GroovySimpleTests.java
diff --git a/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/UnicodeEscapingReader.java b/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/UnicodeEscapingReader.java
@@ -44,6 +44,7 @@ public class UnicodeEscapingReader extends Reader {
 
     private static class DummyLexer extends CharScanner {
         private final Token t = new Token();
+        @Override
         public Token nextToken() throws TokenStreamException {
             return t;
         }
@@ -56,7 +57,7 @@ public int getLine() {
             return 0;
         }
     }
-    
+
     /**
      * Constructor.
      * @param reader The reader that this reader will filter over.
@@ -84,6 +85,7 @@ public void setLexer(CharScanner lexer) {
      * Reads characters from the underlying reader.
      * @see java.io.Reader#read(char[],int,int)
      */
+    @Override
     public int read(char cbuf[], int off, int len) throws IOException {
         int c = 0;
         int count = 0;
@@ -99,9 +101,13 @@ public int read(char cbuf[], int off, int len) throws IOException {
      * translating escapes as required.
      * @see java.io.Reader#close()
      */
+    @Override
     public int read() throws IOException {
         if (hasNextChar) {
             hasNextChar = false;
+            // GRECLIPSE add
+            checkCodePoint(nextChar);
+            // GRECLIPSE add
             write(nextChar);
             return nextChar;
         }
@@ -111,9 +117,12 @@ public int read() throws IOException {
             numUnicodeEscapesFoundOnCurrentLine = 0;
             previousLine = lexer.getLine();
         }
-        
+
         int c = reader.read();
         if (c != '\\') {
+            // GRECLIPSE add
+            checkCodePoint(c);
+            // GRECLIPSE add
             write(c);
             return c;
         }
@@ -147,15 +156,25 @@ public int read() throws IOException {
         }
         int rv = Integer.parseInt(charNum.toString(), 16);
         write(rv);
-        
+
         numUnicodeEscapesFound += 4 + numberOfUChars;
         numUnicodeEscapesFoundOnCurrentLine += 4 + numberOfUChars;
 
         return rv;
     }
+
     private void write(int c) {
         if (sourceBuffer != null) {sourceBuffer.write(c);}
     }
+
+    // GRECLIPSE add
+    private void checkCodePoint(int c) {
+        if (Character.isIdentifierIgnorable(c) || Character.getType(c) == Character.CONTROL && c != '\t' && c != '\r' && c != '\n') {
+            lexer.reportError(String.format("Unexpected character 0x%02X (%s) at column %d", c, Character.getName(c), lexer.getColumn()));
+        }
+    }
+    // GRECLIPSE end
+
     /**
      * Checks that the given character is indeed a hex digit.
      */
@@ -189,6 +208,7 @@ public int getUnescapedUnicodeOffsetCount() {
      *
      * @see java.io.Reader#close()
      */
+    @Override
     public void close() throws IOException {
         reader.close();
     }

diff --git a/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/groovy.g b/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/groovy.g
@@ -3981,10 +3981,15 @@ options {
         return "["+ttn+",\""+t.getText()+"\"]";
     }
 
-    protected GroovyRecognizer parser;  // little-used link; TODO: get rid of
+    protected GroovyRecognizer parser;
+    // GRECLIPSE add
+    public void reportError(String message) {
+        parser.reportError(message, getLine(), getColumn());
+    }
+    // GRECLIPSE end
     private void require(boolean z, String problem, String solution) throws SemanticException {
         // TODO: Direct to a common error handler, rather than through the parser.
-        if (!z && parser!=null)  parser.requireFailed(problem, solution);
+        if (!z && parser!=null) parser.requireFailed(problem, solution);
         if (!z) {
             int lineNum = inputState.getLine(), colNum = inputState.getColumn();
             throw new SemanticException(problem + ";\n   solution: " + solution,
@@ -3993,8 +3998,6 @@ options {
     }
 }
 
-// TODO:  Borneo-style ops.
-
 // OPERATORS
 QUESTION          options {paraphrase="'?'";}           :   '?'             ;
 LPAREN            options {paraphrase="'('";}           :   '('             {++parenLevel;};

diff --git a/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/parser/GroovyLexer.java b/base/org.codehaus.groovy24/src/org/codehaus/groovy/antlr/parser/GroovyLexer.java
@@ -309,10 +309,15 @@ private static String tokenStringOf(Token t) {
         return "["+ttn+",\""+t.getText()+"\"]";
     }
 
-    protected GroovyRecognizer parser;  // little-used link; TODO: get rid of
+    protected GroovyRecognizer parser;
+    // GRECLIPSE add
+    public void reportError(String message) {
+        parser.reportError(message, getLine(), getColumn());
+    }
+    // GRECLIPSE end
     private void require(boolean z, String problem, String solution) throws SemanticException {
         // TODO: Direct to a common error handler, rather than through the parser.
-        if (!z && parser!=null)  parser.requireFailed(problem, solution);
+        if (!z && parser!=null) parser.requireFailed(problem, solution);
         if (!z) {
             int lineNum = inputState.getLine(), colNum = inputState.getColumn();
             throw new SemanticException(problem + ";\n   solution: " + solution,