Fix reading comments with UTF chars (fixes #238) (#240)

gnodet · web-flow · commit 32b72a6e8ff5 · 2023-03-02T08:01:46.000+10:00
* Fix reading comments with UTF chars (fixes #238) * Fix printable methods to account for UTF chars
diff --git a/src/main/java/org/codehaus/plexus/util/xml/pull/MXParser.java b/src/main/java/org/codehaus/plexus/util/xml/pull/MXParser.java
@@ -2981,8 +2981,8 @@ private void parseComment()
         // implements XML 1.0 Section 2.5 Comments
 
         // ASSUMPTION: seen <!-
-        char ch = more();
-        if ( ch != '-' )
+        char cch = more();
+        if ( cch != '-' )
             throw new XmlPullParserException( "expected <!-- for comment start", this, null );
         if ( tokenize )
             posStart = pos;
@@ -2999,7 +2999,19 @@ private void parseComment()
             while ( true )
             {
                 // scan until it hits -->
-                ch = more();
+                cch = more();
+                int ch;
+                char cch2;
+                if ( Character.isHighSurrogate( cch ) )
+                {
+                    cch2 = more();
+                    ch = Character.toCodePoint( cch, cch2 );
+                }
+                else
+                {
+                    cch2 = 0;
+                    ch = cch;
+                }
                 if ( seenDashDash && ch != '>' )
                 {
                     throw new XmlPullParserException( "in comment after two dashes (--) next character must be >"
@@ -3074,7 +3086,11 @@ else if ( ch == '\n' )
                         {
                             if ( pcEnd >= pc.length )
                                 ensurePC( pcEnd );
-                            pc[pcEnd++] = ch;
+                            pc[pcEnd++] = cch;
+                            if ( cch2 != 0 )
+                            {
+                                pc[pcEnd++] = cch2;
+                            }
                         }
                         normalizedCR = false;
                     }
@@ -4153,7 +4169,7 @@ private static boolean isS( char ch )
     // ch != '\u0000' ch < '\uFFFE'
 
     // private char printable(char ch) { return ch; }
-    private static String printable( char ch )
+    private static String printable( int ch )
     {
         if ( ch == '\n' )
         {
@@ -4175,18 +4191,25 @@ else if ( ch == '\'' )
         {
             return "\\u" + Integer.toHexString( ch );
         }
-        return "" + ch;
+        if ( Character.isBmpCodePoint( ch ) )
+        {
+            return Character.toString( ( char ) ch );
+        }
+        else
+        {
+            return new String( new char[] { Character.highSurrogate( ch ), Character.lowSurrogate( ch ) } );
+        }
     }
 
     private static String printable( String s )
     {
         if ( s == null )
             return null;
-        final int sLen = s.length();
+        final int sLen = s.codePointCount(0, s.length());
         StringBuilder buf = new StringBuilder( sLen + 10 );
         for ( int i = 0; i < sLen; ++i )
         {
-            buf.append( printable( s.charAt( i ) ) );
+            buf.append( printable( s.codePointAt( i ) ) );
         }
         s = buf.toString();
         return s;
diff --git a/src/test/java/org/codehaus/plexus/util/xml/pull/MXParserTest.java b/src/test/java/org/codehaus/plexus/util/xml/pull/MXParserTest.java
@@ -1511,4 +1511,30 @@ public void testReplacementInPCArrayWithShorterCharArray()
             fail( "should not raise exception: " + e );
         }
     }
+
+    /**
+     * Ensures emoji can be parsed correctly
+     */
+    @Test
+    public void testUnicode() throws IOException {
+        String input = "<project><!--ALL TEH BOMS!  \uD83D\uDCA3  --></project>";
+
+        try
+        {
+            MXParser parser = new MXParser();
+            parser.setInput( new StringReader( input ) );
+
+            assertEquals( XmlPullParser.START_TAG, parser.nextToken() );
+            assertEquals( "project", parser.getName() );
+            assertEquals( XmlPullParser.COMMENT, parser.nextToken() );
+            assertEquals( "ALL TEH BOMS!  \uD83D\uDCA3  ", parser.getText() );
+            assertEquals( XmlPullParser.END_TAG, parser.nextToken() );
+            assertEquals( "project", parser.getName() );
+        }
+        catch ( XmlPullParserException e )
+        {
+            e.printStackTrace();
+            fail( "should not raise exception: " + e );
+        }
+    }
 }
