Issue checkstyle#3700: Control Characters are not skipped with google…

…_checks config
Luolc · Mar 8, 2017 · 531f7ac · 531f7ac
1 parent 8e64917
commit 531f7ac
Show file tree

Hide file tree

Showing 4 changed files with 65,674 additions and 4 deletions.
diff --git a/...checkstyle/test/chapter2filebasic/rule233nonascii/InputAvoidEscapedUnicodeCharacters.java b/...checkstyle/test/chapter2filebasic/rule233nonascii/InputAvoidEscapedUnicodeCharacters.java
@@ -17,15 +17,15 @@ public Object fooString()
         String fakeUnicode = "asd\tsasd"; 
         String fakeUnicode2 = "\\u23\\u123i\\u";
         String content = "";
-        return "\ufeff" + content ; // byte order mark ok
+        /*byte order mark ok*/return "\ufeff" + content ;
     }
 
     public Object fooChar()
     {
         /*warn*/char unitAbbrev2 = '\u03bc'; 
         char unitAbbrev3 = '\u03bc'; // Greek letter mu, "s" ok
         String content = "";
-        return '\ufeff' + content; // byte order mark ok
+        /*byte order mark ok*/return '\ufeff' + content;
     }
 
     public void multiplyString()

diff --git a/src/main/java/com/puppycrawl/tools/checkstyle/checks/AvoidEscapedUnicodeCharactersCheck.java b/src/main/java/com/puppycrawl/tools/checkstyle/checks/AvoidEscapedUnicodeCharactersCheck.java
@@ -120,8 +120,8 @@ public class AvoidEscapedUnicodeCharactersCheck
 
     /** Regular expression Unicode control characters. */
     private static final Pattern UNICODE_CONTROL = Pattern.compile("\\\\(u|U)"
-            + "(00[0-1][0-1A-Fa-f]|00[8-9][0-9A-Fa-f]|034(f|F)|070(f|F)"
-            + "|180(e|E)|200[b-fB-F]|202[b-eB-E]|206[0-4a-fA-F]"
+            + "(00[0-1][0-9A-Fa-f]|00[8-9][0-9A-Fa-f]|00(a|A)(d|D)|034(f|F)|070(f|F)"
+            + "|180(e|E)|200[b-fB-F]|202[a-eA-E]|206[0-4a-fA-F]"
             + "|[fF]{3}[9a-bA-B]|[fF][eE][fF]{2})");
 
     /** Regular expression for all escaped chars. */

diff --git a/...t/java/com/puppycrawl/tools/checkstyle/checks/AvoidEscapedUnicodeCharactersCheckTest.java b/...t/java/com/puppycrawl/tools/checkstyle/checks/AvoidEscapedUnicodeCharactersCheckTest.java
@@ -24,6 +24,8 @@
 
 import java.io.File;
 import java.io.IOException;
+import java.util.Arrays;
+import java.util.stream.IntStream;
 
 import org.junit.Test;
 
@@ -32,6 +34,109 @@
 import com.puppycrawl.tools.checkstyle.api.TokenTypes;
 
 public class AvoidEscapedUnicodeCharactersCheckTest extends BaseCheckTestSupport {
+
+    private static final int[] C0_CONTROL_CHARACTER_INDICES = {
+        0x0000,
+        0x0001,
+        0x0002,
+        0x0003,
+        0x0004,
+        0x0005,
+        0x0006,
+        0x0007,
+        0x0008,
+        0x0009,
+        0x000a,
+        0x000b,
+        0x000c,
+        0x000d,
+        0x000e,
+        0x000f,
+        0x0010,
+        0x0011,
+        0x0012,
+        0x0013,
+        0x0014,
+        0x0015,
+        0x0016,
+        0x0017,
+        0x0018,
+        0x0019,
+        0x001a,
+        0x001b,
+        0x001c,
+        0x001d,
+        0x001e,
+        0x001f,
+    };
+
+    private static final int[] C1_CONTROL_CHARACTER_INDICES = {
+        0x0080,
+        0x0081,
+        0x0082,
+        0x0083,
+        0x0084,
+        0x0085,
+        0x0086,
+        0x0087,
+        0x0088,
+        0x0089,
+        0x008a,
+        0x008b,
+        0x008c,
+        0x008d,
+        0x008e,
+        0x008f,
+        0x0090,
+        0x0091,
+        0x0092,
+        0x0093,
+        0x0094,
+        0x0095,
+        0x0096,
+        0x0097,
+        0x0098,
+        0x0099,
+        0x009a,
+        0x009b,
+        0x009c,
+        0x009d,
+        0x009e,
+        0x009f,
+    };
+
+    private static final int[] OTHER_CONTROL_CHARACTER_INDICES = {
+        0x00ad,
+        0x034f,
+        0x070f,
+        0x180e,
+        0x200b,
+        0x200c,
+        0x200d,
+        0x200e,
+        0x200f,
+        0x202a,
+        0x202b,
+        0x202c,
+        0x202d,
+        0x202e,
+        0x2060,
+        0x2061,
+        0x2062,
+        0x2063,
+        0x2064,
+        0x206a,
+        0x206b,
+        0x206c,
+        0x206d,
+        0x206e,
+        0x206f,
+        0xfeff,
+        0xfff9,
+        0xfffa,
+        0xfffb,
+    };
+
     @Override
     protected String getPath(String filename) throws IOException {
         return super.getPath("checks" + File.separator + filename);
@@ -224,4 +329,26 @@ public void testGetAcceptableTokens() {
         final int[] expected = {TokenTypes.STRING_LITERAL, TokenTypes.CHAR_LITERAL };
         assertArrayEquals(expected, actual);
     }
+
+    @Test
+    public void testAllowEscapesForControlCharacterSetForAllCharacters() throws Exception {
+        final DefaultConfiguration checkConfig =
+                createCheckConfig(AvoidEscapedUnicodeCharactersCheck.class);
+        checkConfig.addAttribute("allowEscapesForControlCharacters", "true");
+
+        final int indexOfStartLineInInputFile = 6;
+        final String message = getCheckMessage(MSG_KEY);
+        final String[] expected = IntStream.rangeClosed(0, 0xffff)
+                .parallel()
+                .filter(val -> !isControlCharacter(val))
+                .mapToObj(msg -> indexOfStartLineInInputFile + msg + ": " + message)
+                .toArray(String[]::new);
+        verify(checkConfig, getPath("InputAllEscapedUnicodeCharacters.java"), expected);
+    }
+
+    private static boolean isControlCharacter(final int character) {
+        return Arrays.binarySearch(C0_CONTROL_CHARACTER_INDICES, character) >= 0
+                || Arrays.binarySearch(C1_CONTROL_CHARACTER_INDICES, character) >= 0
+                || Arrays.binarySearch(OTHER_CONTROL_CHARACTER_INDICES, character) >= 0;
+    }
 }