Merge "Various regex test fixes."

10 files changed, 134 insertions, 1621 deletions

diff --git a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/MatcherTest.java b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/MatcherTest.java
index bc87439..98450a4 100644
--- a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/MatcherTest.java
+++ b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/MatcherTest.java
@@ -607,16 +607,16 @@ public class MatcherTest extends TestCase {
         assertTrue(matcher.matches());
     }
 
-    public void testAllCodePoints() {
+    public void testAllCodePoints_p() {
         // Regression for HARMONY-3145
         int[] codePoint = new int[1];
         Pattern p = Pattern.compile("(\\p{all})+");
         boolean res = true;
         int cnt = 0;
-        String s;
-        for (int i = 0; i < 0x110000; i++) {
+        int step = 16; // Ideally 1, but devices are still too slow.
+        for (int i = 0; i < 0x110000; i += step) {
             codePoint[0] = i;
-            s = new String(codePoint, 0, 1);
+            String s = new String(codePoint, 0, 1);
             if (!s.matches(p.toString())) {
                 cnt++;
                 res = false;
@@ -624,22 +624,25 @@ public class MatcherTest extends TestCase {
         }
         assertTrue(res);
         assertEquals(0, cnt);
+    }
 
-        p = Pattern.compile("(\\P{all})+");
-        res = true;
-        cnt = 0;
-
-        for (int i = 0; i < 0x110000; i++) {
+    public void testAllCodePoints_P() {
+        // Regression for HARMONY-3145
+        int[] codePoint = new int[1];
+        Pattern p = Pattern.compile("(\\P{all})+");
+        boolean res = true;
+        int cnt = 0;
+        int step = 16; // Ideally 1, but devices are still too slow.
+        for (int i = 0; i < 0x110000; i += step) {
             codePoint[0] = i;
-            s = new String(codePoint, 0, 1);
+            String s = new String(codePoint, 0, 1);
             if (!s.matches(p.toString())) {
                 cnt++;
                 res = false;
             }
         }
-
         assertFalse(res);
-        assertEquals(0x110000, cnt);
+        assertEquals(0x110000 / step, cnt);
     }
 
     /*
@@ -735,14 +738,13 @@ public class MatcherTest extends TestCase {
                 "----1 fish 2 fish red fish 5----", false);
     }
 
-    /*
-     * Test if Matcher's toString conatain pattern information
-     */
     public void testToString() {
-        String result = Pattern.compile("(\\d{1,3})").matcher(
-                "aaaa123456789045").toString();
-        assertTrue("The result doesn't contain pattern info", result
-                .contains("(\\d{1,3})"));
+        Matcher m = Pattern.compile("(\\d{1,3})").matcher("aaaa666456789045");
+        assertEquals("java.util.regex.Matcher[pattern=(\\d{1,3}) region=0,16 lastmatch=]", m.toString());
+        assertTrue(m.find());
+        assertEquals("java.util.regex.Matcher[pattern=(\\d{1,3}) region=0,16 lastmatch=666]", m.toString());
+        m.region(4, 8);
+        assertEquals("java.util.regex.Matcher[pattern=(\\d{1,3}) region=4,8 lastmatch=]", m.toString());
     }
 
     private void hitEndTest(boolean callFind, String testNo, String regex,
diff --git a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/Pattern2Test.java b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/Pattern2Test.java
index e2fec73..cf30fc6 100644
--- a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/Pattern2Test.java
+++ b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/Pattern2Test.java
@@ -287,16 +287,18 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         } catch (PatternSyntaxException e) {
         }
 
+        // icu4c allows 1 to 6 hex digits in \x escapes.
+        p = Pattern.compile("\\xa");
+        p = Pattern.compile("\\xab");
+        p = Pattern.compile("\\xabc");
+        p = Pattern.compile("\\xabcd");
+        p = Pattern.compile("\\xabcde");
+        p = Pattern.compile("\\xabcdef");
+        // (Further digits would just be treated as characters after the escape.)
         try {
-            p = Pattern.compile("\\xa");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\xa;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
+            p = Pattern.compile("\\xg");
+            fail();
+        } catch (PatternSyntaxException expected) {
         }
 
         // Test \0 (octal) sequences (1, 2 and 3 digit)
@@ -369,11 +371,18 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         }
 
         // Test invalid control escapes
-        try {
+        // icu4c 50 accepts this pattern, and treats it as a literal.
+        //try {
             p = Pattern.compile("\\c");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
+            assertTrue(p.matcher("x\\cy").find());
+        //    fail(p.matcher("").toString());
+        //} catch (PatternSyntaxException e) {
+        //}
+
+        // But \cH works.
+        p = Pattern.compile("\\cH");
+        assertTrue(p.matcher("x\u0008y").find());
+        assertFalse(p.matcher("x\\cHy").find());
 
         // originally contributed test did not check the result
         // TODO: check what RI does here
@@ -467,15 +476,18 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         assertTrue(m.matches());
 
         // Test ranges including the "-" character
-        p = Pattern.compile("[\\*-/]_+[---]!+[--AP]");
-        m = p.matcher("-_-!!A");
-        assertTrue(m.matches());
-        m = p.matcher("\u002b_-!!!-");
-        assertTrue(m.matches());
-        m = p.matcher("!_-!@");
-        assertFalse(m.matches());
-        m = p.matcher(",______-!!!!!!!P");
-        assertTrue(m.matches());
+        // "---" collides with icu4c's "--" operator, and likely to be user error anyway.
+        if (false) {
+            p = Pattern.compile("[\\*-/]_+[---]!+[--AP]");
+            m = p.matcher("-_-!!A");
+            assertTrue(m.matches());
+            m = p.matcher("\u002b_-!!!-");
+            assertTrue(m.matches());
+            m = p.matcher("!_-!@");
+            assertFalse(m.matches());
+            m = p.matcher(",______-!!!!!!!P");
+            assertTrue(m.matches());
+        }
 
         // Test nested ranges
         p = Pattern.compile("[pm[t]][a-z]+[[r]lp]");
@@ -503,13 +515,16 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         assertTrue(m.matches());
 
         // Test error cases with &&
-        p = Pattern.compile("[&&[xyz]]");
-        m = p.matcher("&");
-        // System.out.println(m.matches());
-        m = p.matcher("x");
-        // System.out.println(m.matches());
-        m = p.matcher("y");
-        // System.out.println(m.matches());
+        // This is an RI bug that icu4c doesn't have.
+        if (false) {
+            p = Pattern.compile("[&&[xyz]]");
+            m = p.matcher("&");
+            // System.out.println(m.matches());
+            m = p.matcher("x");
+            // System.out.println(m.matches());
+            m = p.matcher("y");
+            // System.out.println(m.matches());
+        }
         p = Pattern.compile("[[xyz]&[axy]]");
         m = p.matcher("x");
         // System.out.println(m.matches());
@@ -521,7 +536,10 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         m = p.matcher("a");
         // System.out.println(m.matches());
 
-        p = Pattern.compile("[[xyz]&&]");
+        // icu4c rightly considers a missing rhs to && a syntax error.
+        if (false) {
+            p = Pattern.compile("[[xyz]&&]");
+        }
 
         p = Pattern.compile("[[abc]&]");
 
@@ -795,54 +813,6 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         // TODO
     }
 
-    public void testUnicodeCategories() throws PatternSyntaxException {
-        // Test Unicode categories using \p and \P
-        // One letter codes: L, M, N, P, S, Z, C
-        // Two letter codes: Lu, Nd, Sc, Sm, ...
-        // See java.lang.Character and Unicode standard for complete list
-        // TODO
-        // Test \p{L}
-        // TODO
-
-        // Test \p{N}
-        // TODO
-
-        // ... etc
-
-        // Test two letter codes:
-        // From unicode.org:
-        // Lu
-        // Ll
-        // Lt
-        // Lm
-        // Lo
-        // Mn
-        // Mc
-        // Me
-        // Nd
-        // Nl
-        // No
-        // Pc
-        // Pd
-        // Ps
-        // Pe
-        // Pi
-        // Pf
-        // Po
-        // Sm
-        // Sc
-        // Sk
-        // So
-        // Zs
-        // Zl
-        // Zp
-        // Cc
-        // Cf
-        // Cs
-        // Co
-        // Cn
-    }
-
     public void testUnicodeBlocks() throws PatternSyntaxException {
         Pattern p;
         Matcher m;
@@ -881,109 +851,34 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
 
             if (UBlocks[i].low > 0) {
                 m = p.matcher(Character.toString((char) (UBlocks[i].low - 1)));
-                assertFalse(m.matches());
+                assertFalse(UBlocks[i].name, m.matches());
             }
             for (j = UBlocks[i].low; j <= UBlocks[i].high; j++) {
                 m = p.matcher(Character.toString((char) j));
-                assertTrue(m.matches());
+                assertTrue(UBlocks[i].name, m.matches());
             }
             if (UBlocks[i].high < 0xFFFF) {
                 m = p.matcher(Character.toString((char) (UBlocks[i].high + 1)));
-                assertFalse(m.matches());
+                assertFalse(UBlocks[i].name, m.matches());
             }
 
             p = Pattern.compile("\\P{In" + UBlocks[i].name + "}");
 
             if (UBlocks[i].low > 0) {
                 m = p.matcher(Character.toString((char) (UBlocks[i].low - 1)));
-                assertTrue(m.matches());
+                assertTrue(UBlocks[i].name, m.matches());
             }
             for (j = UBlocks[i].low; j < UBlocks[i].high; j++) {
                 m = p.matcher(Character.toString((char) j));
-                assertFalse(m.matches());
+                assertFalse(UBlocks[i].name, m.matches());
             }
             if (UBlocks[i].high < 0xFFFF) {
                 m = p.matcher(Character.toString((char) (UBlocks[i].high + 1)));
-                assertTrue(m.matches());
+                assertTrue(UBlocks[i].name, m.matches());
             }
         }
     }
 
-    public void testCapturingGroups() throws PatternSyntaxException {
-        // Test simple capturing groups
-        // TODO
-
-        // Test grouping without capture (?:...)
-        // TODO
-
-        // Test combination of grouping and capture
-        // TODO
-
-        // Test \<num> sequence with capturing and non-capturing groups
-        // TODO
-
-        // Test \<num> with <num> out of range
-        // TODO
-    }
-
-    public void testRepeats() {
-        // Test ?
-        // TODO
-
-        // Test *
-        // TODO
-
-        // Test +
-        // TODO
-
-        // Test {<num>}, including 0, 1 and more
-        // TODO
-
-        // Test {<num>,}, including 0, 1 and more
-        // TODO
-
-        // Test {<n1>,<n2>}, with n1 < n2, n1 = n2 and n1 > n2 (illegal?)
-        // TODO
-    }
-
-    public void testAnchors() throws PatternSyntaxException {
-        // Test ^, default and MULTILINE
-        // TODO
-
-        // Test $, default and MULTILINE
-        // TODO
-
-        // Test \b (word boundary)
-        // TODO
-
-        // Test \B (not a word boundary)
-        // TODO
-
-        // Test \A (beginning of string)
-        // TODO
-
-        // Test \Z (end of string)
-        // TODO
-
-        // Test \z (end of string)
-        // TODO
-
-        // Test \G
-        // TODO
-
-        // Test positive lookahead using (?=...)
-        // TODO
-
-        // Test negative lookahead using (?!...)
-        // TODO
-
-        // Test positive lookbehind using (?<=...)
-        // TODO
-
-        // Test negative lookbehind using (?<!...)
-        // TODO
-    }
-
     public void testMisc() throws PatternSyntaxException {
         Pattern p;
         Matcher m;
@@ -1127,11 +1022,9 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
         assertFalse(m.find());
 
         m = p.matcher("");
-        // FIXME: This matches the reference behaviour but is
-        // inconsistent with matching "a" - ie. the end of the
-        // target string should match against $ always but this
-        // appears to work with the null string only when not in
-        // multiline mode (see below)
+        // This differs from the RI behaviour but seems more correct.
+        assertTrue(m.find());
+        assertTrue(m.group().equals(""));
         assertFalse(m.find());
 
         p = Pattern.compile("^.*$");
@@ -1371,7 +1264,7 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
             /* 3300; 33FF; CJK Compatibility */
             new UBInfo(0x3300, 0x33FF, "CJKCompatibility"), // Character.UnicodeBlock.CJK_COMPATIBILITY
             /* 3400; 4DB5; CJK Unified Ideographs Extension A */
-            new UBInfo(0x3400, 0x4DB5, "CJKUnifiedIdeographsExtensionA"), // Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A
+            new UBInfo(0x3400, 0x4DBF, "CJKUnifiedIdeographsExtensionA"), // Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A
             /* 4E00; 9FFF; CJK Unified Ideographs */
             new UBInfo(0x4E00, 0x9FFF, "CJKUnifiedIdeographs"), // Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS
             /* A000; A48F; Yi Syllables */
@@ -1379,7 +1272,7 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
             /* A490; A4CF; Yi Radicals */
             new UBInfo(0xA490, 0xA4CF, "YiRadicals"), // Character.UnicodeBlock.YI_RADICALS
             /* AC00; D7A3; Hangul Syllables */
-            new UBInfo(0xAC00, 0xD7A3, "HangulSyllables"), // Character.UnicodeBlock.HANGUL_SYLLABLES
+            new UBInfo(0xAC00, 0xD7AF, "HangulSyllables"), // Character.UnicodeBlock.HANGUL_SYLLABLES
             /* D800; DB7F; High Surrogates */
             /* DB80; DBFF; High Private Use Surrogates */
             /* DC00; DFFF; Low Surrogates */
@@ -1397,13 +1290,10 @@ p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
             /* FE50; FE6F; Small Form Variants */
             new UBInfo(0xFE50, 0xFE6F, "SmallFormVariants"), // Character.UnicodeBlock.SMALL_FORM_VARIANTS
             /* FE70; FEFE; Arabic Presentation Forms-B */
-            // new UBInfo (0xFE70,0xFEFE,"InArabicPresentationForms-B"), //
-            // Character.UnicodeBlock.ARABIC_PRESENTATION_FORMS_B
-            /* FEFF; FEFF; Specials */
-            new UBInfo(0xFEFF, 0xFEFF, "Specials"), // Character.UnicodeBlock.SPECIALS
+            new UBInfo(0xFE70, 0xFEFF, "ArabicPresentationForms-B"), // Character.UnicodeBlock.ARABIC_PRESENTATION_FORMS_B
             /* FF00; FFEF; Halfwidth and Fullwidth Forms */
             new UBInfo(0xFF00, 0xFFEF, "HalfwidthandFullwidthForms"), // Character.UnicodeBlock.HALFWIDTH_AND_FULLWIDTH_FORMS
             /* FFF0; FFFD; Specials */
-            new UBInfo(0xFFF0, 0xFFFD, "Specials") // Character.UnicodeBlock.SPECIALS
+            new UBInfo(0xFFF0, 0xFFFF, "Specials") // Character.UnicodeBlock.SPECIALS
     };
-}
\ No newline at end of file
+}
diff --git a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternErrorTest.java b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternErrorTest.java
index a8eef6d..4abb2a0 100644
--- a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternErrorTest.java
+++ b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternErrorTest.java
@@ -4,9 +4,9 @@
  * The ASF licenses this file to You under the Apache License, Version 2.0
  * (the "License"); you may not use this file except in compliance with
  * the License.  You may obtain a copy of the License at
- * 
+ *
  *     http://www.apache.org/licenses/LICENSE-2.0
- * 
+ *
  * Unless required by applicable law or agreed to in writing, software
  * distributed under the License is distributed on an "AS IS" BASIS,
  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
@@ -50,11 +50,13 @@ public class PatternErrorTest extends TestCase {
         flags |= Pattern.COMMENTS;
         flags |= Pattern.DOTALL;
         flags |= Pattern.UNICODE_CASE;
+        flags &= ~Pattern.CANON_EQ; // Android always throws given this flag.
         Pattern.compile("foo", flags);
 
         // add invalid flags - should get IllegalArgumentException
         // regression test for HARMONY-4248
         flags |= 0xFFFFFFFF;
+        flags &= ~Pattern.CANON_EQ; // Android always throws given this flag.
         try {
             Pattern.compile("foo", flags);
             fail("Expected IllegalArgumentException to be thrown");
diff --git a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternTest.java b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternTest.java
index 33da926..9977d91 100644
--- a/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternTest.java
+++ b/harmony-tests/src/test/java/org/apache/harmony/tests/java/util/regex/PatternTest.java
@@ -277,6 +277,12 @@ public class PatternTest extends TestCase {
     }
 
     public void testFlagsMethod() {
+        // icu4c doesn't count inline flags that span the entire regex as being global flags.
+        // Android just returns those flags actually passed to Pattern.compile.
+        if (true) {
+            return;
+        }
+
         String baseString;
         Pattern pat;
 
@@ -507,16 +513,22 @@ public class PatternTest extends TestCase {
         assertTrue(mat.matches());
     }
 
-    public void testRegressions() {
+    public void test_bug_181() {
         // Bug 181
         Pattern.compile("[\\t-\\r]");
+    }
 
+    public void test_bug_4472() {
         // HARMONY-4472
         Pattern.compile("a*.+");
+    }
 
+    public void test_bug_187() {
         // Bug187
-        Pattern
-                .compile("|(?idmsux-idmsux)|(?idmsux-idmsux)|[^|\\[-\\0274|\\,-\\\\[^|W\\}\\nq\\x65\\002\\xFE\\05\\06\\00\\x66\\x47i\\,\\xF2\\=\\06\\u0EA4\\x9B\\x3C\\f\\|\\{\\xE5\\05\\r\\u944A\\xCA\\e|\\x19\\04\\x07\\04\\u607B\\023\\0073\\x91Tr\\0150\\x83]]?(?idmsux-idmsux:\\p{Alpha}{7}?)||(?<=[^\\uEC47\\01\\02\\u3421\\a\\f\\a\\013q\\035w\\e])(?<=\\p{Punct}{0,}?)(?=^\\p{Lower})(?!\\b{8,14})(?<![|\\00-\\0146[^|\\04\\01\\04\\060\\f\\u224DO\\x1A\\xC4\\00\\02\\0315\\0351\\u84A8\\xCBt\\xCC\\06|\\0141\\00\\=\\e\\f\\x6B\\0026Tb\\040\\x76xJ&&[\\\\-\\]\\05\\07\\02\\u2DAF\\t\\x9C\\e\\0023\\02\\,X\\e|\\u6058flY\\u954C]]]{5}?)(?<=\\p{Sc}{8}+)[^|\\026-\\u89BA|o\\u6277\\t\\07\\x50&&\\p{Punct}]{8,14}+((?<=^\\p{Punct})|(?idmsux-idmsux)||(?>[\\x3E-\\]])|(?idmsux-idmsux:\\p{Punct})|(?<![\\0111\\0371\\xDF\\u6A49\\07\\u2A4D\\00\\0212\\02Xd-\\xED[^\\a-\\0061|\\0257\\04\\f\\[\\0266\\043\\03\\x2D\\042&&[^\\f-\\]&&\\s]]])|(?>[|\\n\\042\\uB09F\\06\\u0F2B\\uC96D\\x89\\uC166\\xAA|\\04-\\][^|\\a\\|\\rx\\04\\uA770\\n\\02\\t\\052\\056\\0274\\|\\=\\07\\e|\\00-\\x1D&&[^\\005\\uB15B\\uCDAC\\n\\x74\\0103\\0147\\uD91B\\n\\062G\\u9B4B\\077\\}\\0324&&[^\\0302\\,\\0221\\04\\u6D16\\04xy\\uD193\\[\\061\\06\\045\\x0F|\\e\\xBB\\f\\u1B52\\023\\u3AD2\\033\\007\\022\\}\\x66\\uA63FJ-\\0304]]]]{0,0})||(?<![^|\\0154U\\u0877\\03\\fy\\n\\|\\0147\\07-\\=[|q\\u69BE\\0243\\rp\\053\\02\\x33I\\u5E39\\u9C40\\052-\\xBC[|\\0064-\\?|\\uFC0C\\x30\\0060\\x45\\\\\\02\\?p\\xD8\\0155\\07\\0367\\04\\uF07B\\000J[^|\\0051-\\{|\\u9E4E\\u7328\\]\\u6AB8\\06\\x71\\a\\]\\e\\|KN\\u06AA\\0000\\063\\u2523&&[\\005\\0277\\x41U\\034\\}R\\u14C7\\u4767\\x09\\n\\054Ev\\0144\\<\\f\\,Q-\\xE4]]]]]{3}+)|(?>^+)|(?![^|\\|\\nJ\\t\\<\\04E\\\\\\t\\01\\\\\\02\\|\\=\\}\\xF3\\uBEC2\\032K\\014\\uCC5F\\072q\\|\\0153\\xD9\\0322\\uC6C8[^\\t\\0342\\x34\\x91\\06\\{\\xF1\\a\\u1710\\?\\xE7\\uC106\\02pF\\<&&[^|\\]\\064\\u381D\\u50CF\\eO&&[^|\\06\\x2F\\04\\045\\032\\u8536W\\0377\\0017|\\x06\\uE5FA\\05\\xD4\\020\\04c\\xFC\\02H\\x0A\\r]]]]+?)(?idmsux-idmsux)|(?<![|\\r-\\,&&[I\\t\\r\\0201\\xDB\\e&&[^|\\02\\06\\00\\<\\a\\u7952\\064\\051\\073\\x41\\?n\\040\\0053\\031&&[\\x15-\\|]]]]{8,11}?)(?![^|\\<-\\uA74B\\xFA\\u7CD2\\024\\07n\\<\\x6A\\0042\\uE4FF\\r\\u896B\\[\\=\\042Y&&^\\p{ASCII}]++)|(?<![R-\\|&&[\\a\\0120A\\u6145\\<\\050-d[|\\e-\\uA07C|\\016-\\u80D9]]]{1,}+)|(?idmsux-idmsux)|(?idmsux-idmsux)|(?idmsux-idmsux:\\B{6,}?)|(?<=\\D{5,8}?)|(?>[\\{-\\0207|\\06-\\0276\\p{XDigit}])(?idmsux-idmsux:[^|\\x52\\0012\\]u\\xAD\\0051f\\0142\\\\l\\|\\050\\05\\f\\t\\u7B91\\r\\u7763\\{|h\\0104\\a\\f\\0234\\u2D4F&&^\\P{InGreek}]))");
+        Pattern.compile("|(?idmsux-idmsux)|(?idmsux-idmsux)|[^|\\[-\\0274|\\,-\\\\[^|W\\}\\nq\\x65\\002\\xFE\\05\\06\\00\\x66\\x47i\\,\\xF2\\=\\06\\u0EA4\\x9B\\x3C\\f\\|\\{\\xE5\\05\\r\\u944A\\xCA\\e|\\x19\\04\\x07\\04\\u607B\\023\\0073\\x91Tr\\0150\\x83]]?(?idmsux-idmsux:\\p{Alpha}{7}?)||(?<=[^\\uEC47\\01\\02\\u3421\\a\\f\\a\\013q\\035w\\e])(?<=\\p{Punct}{0,}?)(?=^\\p{Lower})(?!\\b{8,14})(?<![|\\00-\\0146[^|\\04\\01\\04\\060\\f\\u224DO\\x1A\\xC4\\00\\02\\0315\\0351\\u84A8\\xCBt\\xCC\\06|\\0141\\00\\=\\e\\f\\x6B\\0026Tb\\040\\x76xJ&&[\\\\-\\]\\05\\07\\02\\u2DAF\\t\\x9C\\e\\0023\\02\\,X\\e|\\u6058flY\\u954C]]]{5}?)(?<=\\p{Sc}{8}+)[^|\\026-\\u89BA|o\\u6277\\t\\07\\x50&&\\p{Punct}]{8,14}+((?<=^\\p{Punct})|(?idmsux-idmsux)||(?>[\\x3E-\\]])|(?idmsux-idmsux:\\p{Punct})|(?<![\\0111\\0371\\xDF\\u6A49\\07\\u2A4D\\00\\0212\\02Xd-\\xED[^\\a-\\0061|\\0257\\04\\f\\[\\0266\\043\\03\\x2D\\042&&[^\\f-\\]&&\\s]]])|(?>[|\\n\\042\\uB09F\\06\\u0F2B\\uC96D\\x89\\uC166\\xAA|\\04-\\][^|\\a\\|\\rx\\04\\uA770\\n\\02\\t\\052\\056\\0274\\|\\=\\07\\e|\\00-\\x1D&&[^\\005\\uB15B\\uCDAC\\n\\x74\\0103\\0147\\uD91B\\n\\062G\\u9B4B\\077\\}\\0324&&[^\\0302\\,\\0221\\04\\u6D16\\04xy\\uD193\\[\\061\\06\\045\\x0F|\\e\\xBB\\f\\u1B52\\023\\u3AD2\\033\\007\\022\\}\\x66\\uA63FJ-\\0304]]]]{0,0})||(?<![^|\\0154U\\u0877\\03\\fy\\n\\|\\0147\\07-\\=[|q\\u69BE\\0243\\rp\\053\\02\\x33I\\u5E39\\u9C40\\052-\\xBC[|\\0064-\\?|\\uFC0C\\x30\\0060\\x45\\\\\\02\\?p\\xD8\\0155\\07\\0367\\04\\uF07B\\000J[^|\\0051-\\{|\\u9E4E\\u7328\\]\\u6AB8\\06\\x71\\a\\]\\e\\|KN\\u06AA\\0000\\063\\u2523&&[\\005\\0277\\x41U\\034\\}R\\u14C7\\u4767\\x09\\n\\054Ev\\0144\\<\\f\\,Q-\\xE4]]]]]{3}+)|(?>^+)|(?![^|\\|\\nJ\\t\\<\\04E\\\\\\t\\01\\\\\\02\\|\\=\\}\\xF3\\uBEC2\\032K\\014\\uCC5F\\072q\\|\\0153\\xD9\\0322\\uC6C8[^\\t\\0342\\x34\\x91\\06\\{\\xF1\\a\\u1710\\?\\xE7\\uC106\\02pF\\<&&[^|\\]\\064\\u381D\\u50CF\\eO&&[^|\\06\\x2F\\04\\045\\032\\u8536W\\0377\\0017|\\x06\\uE5FA\\05\\xD4\\020\\04c\\xFC\\02H\\x0A\\r]]]]+?)(?idmsux-idmsux)|(?<![|\\r-\\,&&[I\\t\\r\\0201\\xDB\\e&&[^|\\02\\06\\00\\<\\a\\u7952\\064\\051\\073\\x41\\?n\\040\\0053\\031&&[\\x15-\\|]]]]{8,11}?)(?![^|\\<-\\uA74B\\xFA\\u7CD2\\024\\07n\\<\\x6A\\0042\\uE4FF\\r\\u896B\\[\\=\\042Y&&^\\p{ASCII}]++)|(?<![R-\\|&&[\\a\\0120A\\u6145\\<\\050-d[|\\e-\\uA07C|\\016-\\u80D9]]]{1,}+)|(?idmsux-idmsux)|(?idmsux-idmsux)|(?idmsux-idmsux:\\B{6,}?)|(?<=\\D{5,8}?)|(?>[\\{-\\0207|\\06-\\0276\\p{XDigit}])(?idmsux-idmsux:[^|\\x52\\0012\\]u\\xAD\\0051f\\0142\\\\l\\|\\050\\05\\f\\t\\u7B91\\r\\u7763\\{|h\\0104\\a\\f\\0234\\u2D4F&&^\\P{InGreek}]))");
+    }
+
+    public void test_bug_5858() {
         // HARMONY-5858
         Pattern.compile("\\u6211", Pattern.LITERAL);
     }
@@ -940,13 +952,11 @@ public class PatternTest extends TestCase {
         boolean isCompiled = false;
 
         try {
-            Pattern.compile("(?:)", Pattern.CANON_EQ);
-            Pattern.compile("(?:)", Pattern.CANON_EQ | Pattern.DOTALL);
-            Pattern
-                    .compile("(?:)", Pattern.CANON_EQ
-                            | Pattern.CASE_INSENSITIVE);
-            Pattern.compile("(?:)", Pattern.CANON_EQ | Pattern.COMMENTS
-                    | Pattern.UNIX_LINES);
+            // icu4c doesn't support CANON_EQ.
+            Pattern.compile("(?:)"/*, Pattern.CANON_EQ*/);
+            Pattern.compile("(?:)", /*Pattern.CANON_EQ |*/ Pattern.DOTALL);
+            Pattern.compile("(?:)", /*Pattern.CANON_EQ |*/ Pattern.CASE_INSENSITIVE);
+            Pattern.compile("(?:)", /*Pattern.CANON_EQ |*/ Pattern.COMMENTS | Pattern.UNIX_LINES);
             isCompiled = true;
         } catch (PatternSyntaxException e) {
             System.out.println(e);
@@ -1055,6 +1065,10 @@ public class PatternTest extends TestCase {
     }
 
     public void testCanonEqFlag() {
+        // icu4c doesn't support CANON_EQ.
+        if (true) {
+            return;
+        }
 
         /*
          * for decompositions see
@@ -1251,6 +1265,11 @@ public class PatternTest extends TestCase {
     }
 
     public void testIndexesCanonicalEq() {
+        // icu4c doesn't support CANON_EQ.
+        if (true) {
+            return;
+        }
+
         String baseString;
         String testString;
         Pattern pat;
@@ -1280,6 +1299,10 @@ public class PatternTest extends TestCase {
     }
 
     public void testCanonEqFlagWithSupplementaryCharacters() {
+        // icu4c doesn't support CANON_EQ.
+        if (true) {
+            return;
+        }
 
         /*
          * \u1D1BF->\u1D1BB\u1D16F->\u1D1B9\u1D165\u1D16F in UTF32
diff --git a/luni/src/main/java/java/util/regex/Matcher.java b/luni/src/main/java/java/util/regex/Matcher.java
index 320e14c..d58d092 100644
--- a/luni/src/main/java/java/util/regex/Matcher.java
+++ b/luni/src/main/java/java/util/regex/Matcher.java
@@ -656,6 +656,16 @@ public final class Matcher implements MatchResult {
         }
     }
 
+    /**
+     * Returns a string representing this {@code Matcher}.
+     * The format of this string is unspecified.
+     */
+    @Override public String toString() {
+        return getClass().getName() + "[pattern=" + pattern() +
+            " region=" + regionStart() + "," + regionEnd() +
+            " lastmatch=" + (matchFound ? group() : "") + "]";
+    }
+
     private static native void closeImpl(long addr);
     private static native boolean findImpl(long addr, String s, int startIndex, int[] offsets);
     private static native boolean findNextImpl(long addr, String s, int[] offsets);
diff --git a/luni/src/main/java/java/util/regex/Pattern.java b/luni/src/main/java/java/util/regex/Pattern.java
index 44b749e..45bd800 100644
--- a/luni/src/main/java/java/util/regex/Pattern.java
+++ b/luni/src/main/java/java/util/regex/Pattern.java
@@ -385,6 +385,10 @@ public final class Pattern implements Serializable {
         if ((flags & CANON_EQ) != 0) {
             throw new UnsupportedOperationException("CANON_EQ flag not supported");
         }
+        int supportedFlags = CASE_INSENSITIVE | COMMENTS | DOTALL | LITERAL | MULTILINE | UNICODE_CASE | UNIX_LINES;
+        if ((flags & ~supportedFlags) != 0) {
+            throw new IllegalArgumentException("Unsupported flags: " + (flags & ~supportedFlags));
+        }
         this.pattern = pattern;
         this.flags = flags;
         compile();
diff --git a/luni/src/test/java/libcore/java/util/regex/OldMatcherTest.java b/luni/src/test/java/libcore/java/util/regex/OldMatcherTest.java
index 8d6e186..deb0626 100644
--- a/luni/src/test/java/libcore/java/util/regex/OldMatcherTest.java
+++ b/luni/src/test/java/libcore/java/util/regex/OldMatcherTest.java
@@ -421,44 +421,6 @@ public class OldMatcherTest extends TestCase {
 
     }
 
-    // BEGIN android-note
-    // Test took ages, now going in steps of 16 code points to speed things up.
-    // END android-note
-    public void testAllCodePoints() {
-        // Regression for HARMONY-3145
-        int[] codePoint = new int[1];
-        Pattern p = Pattern.compile("(\\p{all})+");
-        boolean res = true;
-        int cnt = 0;
-        String s;
-        for (int i = 0; i < 0x110000; i = i + 0x10) {
-            codePoint[0] = i;
-            s = new String(codePoint, 0, 1);
-            if (!s.matches(p.toString())) {
-                cnt++;
-                res = false;
-            }
-        }
-        assertTrue(res);
-        assertEquals(0, cnt);
-
-        p = Pattern.compile("(\\P{all})+");
-        res = true;
-        cnt = 0;
-
-        for (int i = 0; i < 0x110000; i = i + 0x10) {
-            codePoint[0] = i;
-            s = new String(codePoint, 0, 1);
-            if (!s.matches(p.toString())) {
-                cnt++;
-                res = false;
-            }
-        }
-
-        assertFalse(res);
-        assertEquals(0x110000 / 0x10, cnt);
-    }
-
     public void test_regionStart() {
         String testPattern = "(abb)";
         String testString = "cccabbabbabbabbabb";
diff --git a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/Pattern2Test.java b/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/Pattern2Test.java
index 389194a..ee8ecfa 100644
--- a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/Pattern2Test.java
+++ b/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/Pattern2Test.java
@@ -28,784 +28,6 @@ import junit.framework.TestCase;
  */
 public class Pattern2Test extends TestCase {
 
-    public void testSimpleMatch() throws PatternSyntaxException {
-        Pattern p = Pattern.compile("foo.*");
-
-        Matcher m1 = p.matcher("foo123");
-        assertTrue(m1.matches());
-        assertTrue(m1.find(0));
-        assertTrue(m1.lookingAt());
-
-        Matcher m2 = p.matcher("fox");
-        assertFalse(m2.matches());
-        assertFalse(m2.find(0));
-        assertFalse(m2.lookingAt());
-
-        assertTrue(Pattern.matches("foo.*", "foo123"));
-        assertFalse(Pattern.matches("foo.*", "fox"));
-
-        assertFalse(Pattern.matches("bar", "foobar"));
-
-        assertTrue(Pattern.matches("", ""));
-    }
-    public void testCursors() {
-        Pattern p;
-        Matcher m;
-
-        try {
-            p = Pattern.compile("foo");
-
-            m = p.matcher("foobar");
-            assertTrue(m.find());
-            assertEquals(0, m.start());
-            assertEquals(3, m.end());
-            assertFalse(m.find());
-
-            // Note: also testing reset here
-            m.reset();
-            assertTrue(m.find());
-            assertEquals(0, m.start());
-            assertEquals(3, m.end());
-            assertFalse(m.find());
-
-            m.reset("barfoobar");
-            assertTrue(m.find());
-            assertEquals(3, m.start());
-            assertEquals(6, m.end());
-            assertFalse(m.find());
-
-            m.reset("barfoo");
-            assertTrue(m.find());
-            assertEquals(3, m.start());
-            assertEquals(6, m.end());
-            assertFalse(m.find());
-
-            m.reset("foobarfoobarfoo");
-            assertTrue(m.find());
-            assertEquals(0, m.start());
-            assertEquals(3, m.end());
-            assertTrue(m.find());
-            assertEquals(6, m.start());
-            assertEquals(9, m.end());
-            assertTrue(m.find());
-            assertEquals(12, m.start());
-            assertEquals(15, m.end());
-            assertFalse(m.find());
-            assertTrue(m.find(0));
-            assertEquals(0, m.start());
-            assertEquals(3, m.end());
-            assertTrue(m.find(4));
-            assertEquals(6, m.start());
-            assertEquals(9, m.end());
-        } catch (PatternSyntaxException e) {
-            System.out.println(e.getMessage());
-            fail();
-        }
-    }
-    public void testGroups() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        p = Pattern.compile("(p[0-9]*)#?(q[0-9]*)");
-
-        m = p.matcher("p1#q3p2q42p5p71p63#q888");
-        assertTrue(m.find());
-        assertEquals(0, m.start());
-        assertEquals(5, m.end());
-        assertEquals(2, m.groupCount());
-        assertEquals(0, m.start(0));
-        assertEquals(5, m.end(0));
-        assertEquals(0, m.start(1));
-        assertEquals(2, m.end(1));
-        assertEquals(3, m.start(2));
-        assertEquals(5, m.end(2));
-        assertEquals("p1#q3", m.group());
-        assertEquals("p1#q3", m.group(0));
-        assertEquals("p1", m.group(1));
-        assertEquals("q3", m.group(2));
-
-        assertTrue(m.find());
-        assertEquals(5, m.start());
-        assertEquals(10, m.end());
-        assertEquals(2, m.groupCount());
-        assertEquals(10, m.end(0));
-        assertEquals(5, m.start(1));
-        assertEquals(7, m.end(1));
-        assertEquals(7, m.start(2));
-        assertEquals(10, m.end(2));
-        assertEquals("p2q42", m.group());
-        assertEquals("p2q42", m.group(0));
-        assertEquals("p2", m.group(1));
-        assertEquals("q42", m.group(2));
-
-        assertTrue(m.find());
-        assertEquals(15, m.start());
-        assertEquals(23, m.end());
-        assertEquals(2, m.groupCount());
-        assertEquals(15, m.start(0));
-        assertEquals(23, m.end(0));
-        assertEquals(15, m.start(1));
-        assertEquals(18, m.end(1));
-        assertEquals(19, m.start(2));
-        assertEquals(23, m.end(2));
-        assertEquals("p63#q888", m.group());
-        assertEquals("p63#q888", m.group(0));
-        assertEquals("p63", m.group(1));
-        assertEquals("q888", m.group(2));
-        assertFalse(m.find());
-    }
-
-    public void testReplace() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        // Note: examples from book,
-        // Hitchens, Ron, 2002, "Java NIO", O'Reilly, page 171
-        p = Pattern.compile("a*b");
-
-        m = p.matcher("aabfooaabfooabfoob");
-        assertTrue(m.replaceAll("-").equals("-foo-foo-foo-"));
-        assertTrue(m.replaceFirst("-").equals("-fooaabfooabfoob"));
-
-        /*
-         * p = Pattern.compile ("\\p{Blank}");
-         *
-         * m = p.matcher ("fee fie foe fum"); assertTrue
-         * (m.replaceFirst("-").equals ("fee-fie foe fum")); assertTrue
-         * (m.replaceAll("-").equals ("fee-fie-foe-fum"));
-         */
-
-        p = Pattern.compile("([bB])yte");
-
-        m = p.matcher("Byte for byte");
-        assertTrue(m.replaceFirst("$1ite").equals("Bite for byte"));
-        assertTrue(m.replaceAll("$1ite").equals("Bite for bite"));
-
-        p = Pattern.compile("\\d\\d\\d\\d([- ])");
-
-        m = p.matcher("card #1234-5678-1234");
-        assertTrue(m.replaceFirst("xxxx$1").equals("card #xxxx-5678-1234"));
-        assertTrue(m.replaceAll("xxxx$1").equals("card #xxxx-xxxx-1234"));
-
-        p = Pattern.compile("(up|left)( *)(right|down)");
-
-        m = p.matcher("left right, up down");
-        assertTrue(m.replaceFirst("$3$2$1").equals("right left, up down"));
-        assertTrue(m.replaceAll("$3$2$1").equals("right left, down up"));
-
-        p = Pattern.compile("([CcPp][hl]e[ea]se)");
-
-        m = p.matcher("I want cheese. Please.");
-        assertTrue(m.replaceFirst("<b> $1 </b>").equals(
-                "I want <b> cheese </b>. Please."));
-        assertTrue(m.replaceAll("<b> $1 </b>").equals(
-                "I want <b> cheese </b>. <b> Please </b>."));
-    }
-
-    public void testEscapes() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        // Test \\ sequence
-        p = Pattern.compile("([a-z]+)\\\\([a-z]+);");
-        m = p.matcher("fred\\ginger;abbott\\costello;jekell\\hyde;");
-        assertTrue(m.find());
-        assertEquals("fred", m.group(1));
-        assertEquals("ginger", m.group(2));
-        assertTrue(m.find());
-        assertEquals("abbott", m.group(1));
-        assertEquals("costello", m.group(2));
-        assertTrue(m.find());
-        assertEquals("jekell", m.group(1));
-        assertEquals("hyde", m.group(2));
-        assertFalse(m.find());
-
-        // Test \n, \t, \r, \f, \e, \a sequences
-        p = Pattern.compile("([a-z]+)[\\n\\t\\r\\f\\e\\a]+([a-z]+)");
-        m = p.matcher("aa\nbb;cc\u0009\rdd;ee\u000C\u001Bff;gg\n\u0007hh");
-        assertTrue(m.find());
-        assertEquals("aa", m.group(1));
-        assertEquals("bb", m.group(2));
-        assertTrue(m.find());
-        assertEquals("cc", m.group(1));
-        assertEquals("dd", m.group(2));
-        assertTrue(m.find());
-        assertEquals("ee", m.group(1));
-        assertEquals("ff", m.group(2));
-        assertTrue(m.find());
-        assertEquals("gg", m.group(1));
-        assertEquals("hh", m.group(2));
-        assertFalse(m.find());
-
-        // Test \\u and \\x sequences
-/*        p = Pattern.compile("([0-9]+)[\\u0020:\\x21];");
-        m = p.matcher("11:;22 ;33-;44!;");
-        assertTrue(m.find());
-        assertEquals("11", m.group(1));
-        assertTrue(m.find());
-        assertEquals("22", m.group(1));
-        assertTrue(m.find());
-        assertEquals("44", m.group(1));
-        assertFalse(m.find());
-*/
-        // Test invalid unicode sequences
-/*        try {
-            p = Pattern.compile("\\u");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\u;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\u002");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\u002;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        // Test invalid hex sequences
-        try {
-            p = Pattern.compile("\\x");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\x;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\xa");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\xa;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-*/
-        // Test \0 (octal) sequences (1, 2 and 3 digit)
-        p = Pattern.compile("([0-9]+)[\\07\\040\\0160];");
-        m = p.matcher("11\u0007;22:;33 ;44p;");
-        assertTrue(m.find());
-        assertEquals("11", m.group(1));
-        assertTrue(m.find());
-        assertEquals("33", m.group(1));
-        assertTrue(m.find());
-        assertEquals("44", m.group(1));
-        assertFalse(m.find());
-
-        // Test invalid octal sequences
-        try {
-            p = Pattern.compile("\\08");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        //originally contributed test did not check the result
-        //TODO: check what RI does here
-//        try {
-//            p = Pattern.compile("\\0477");
-//            fail("PatternSyntaxException expected");
-//        } catch (PatternSyntaxException e) {
-//        }
-
-        try {
-            p = Pattern.compile("\\0");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\0;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-
-        // Test \c (control character) sequence
-        p = Pattern.compile("([0-9]+)[\\cA\\cB\\cC\\cD];");
-        m = p.matcher("11\u0001;22:;33\u0002;44p;55\u0003;66\u0004;");
-        assertTrue(m.find());
-        assertEquals("11", m.group(1));
-        assertTrue(m.find());
-        assertEquals("33", m.group(1));
-        assertTrue(m.find());
-        assertEquals("55", m.group(1));
-        assertTrue(m.find());
-        assertEquals("66", m.group(1));
-        assertFalse(m.find());
-
-        // More thorough control escape test
-        // Ensure that each escape matches exactly the corresponding
-        // character
-        // code and no others (well, from 0-255 at least)
-        int i, j;
-        for (i = 0; i < 26; i++) {
-            p = Pattern.compile("\\c" + Character.toString((char) ('A' + i)));
-            int match_char = -1;
-            for (j = 0; j < 255; j++) {
-                m = p.matcher(Character.toString((char) j));
-                if (m.matches()) {
-                    assertEquals(-1, match_char);
-                    match_char = j;
-                }
-            }
-            assertTrue(match_char == i + 1);
-        }
-
-        // Test invalid control escapes
-// BEGIN android-removed
-// ICU doesn't complain about illegal control sequences
-//        try {
-//            p = Pattern.compile("\\c");
-//            fail("PatternSyntaxException expected");
-//        } catch (PatternSyntaxException e) {
-//        }
-// END android-removed
-
-        //originally contributed test did not check the result
-        //TODO: check what RI does here
-//        try {
-//            p = Pattern.compile("\\c;");
-//            fail("PatternSyntaxException expected");
-//        } catch (PatternSyntaxException e) {
-//        }
-//
-//        try {
-//            p = Pattern.compile("\\ca;");
-//            fail("PatternSyntaxException expected");
-//        } catch (PatternSyntaxException e) {
-//        }
-//
-//        try {
-//            p = Pattern.compile("\\c4;");
-//            fail("PatternSyntaxException expected");
-//        } catch (PatternSyntaxException e) {
-//        }
-    }
-    public void testCharacterClasses() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        // Test one character range
-        p = Pattern.compile("[p].*[l]");
-        m = p.matcher("paul");
-        assertTrue(m.matches());
-        m = p.matcher("pool");
-        assertTrue(m.matches());
-        m = p.matcher("pong");
-        assertFalse(m.matches());
-        m = p.matcher("pl");
-        assertTrue(m.matches());
-
-        // Test two character range
-        p = Pattern.compile("[pm].*[lp]");
-        m = p.matcher("prop");
-        assertTrue(m.matches());
-        m = p.matcher("mall");
-        assertTrue(m.matches());
-        m = p.matcher("pong");
-        assertFalse(m.matches());
-        m = p.matcher("pill");
-        assertTrue(m.matches());
-
-        // Test range including [ and ]
-        p = Pattern.compile("[<\\[].*[\\]>]");
-        m = p.matcher("<foo>");
-        assertTrue(m.matches());
-        m = p.matcher("[bar]");
-        assertTrue(m.matches());
-        m = p.matcher("{foobar]");
-        assertFalse(m.matches());
-        m = p.matcher("<pill]");
-        assertTrue(m.matches());
-
-        // Test range using ^
-        p = Pattern.compile("[^bc][a-z]+[tr]");
-        m = p.matcher("pat");
-        assertTrue(m.matches());
-        m = p.matcher("liar");
-        assertTrue(m.matches());
-        m = p.matcher("car");
-        assertFalse(m.matches());
-        m = p.matcher("gnat");
-        assertTrue(m.matches());
-
-        // Test character range using -
-        p = Pattern.compile("[a-z]_+[a-zA-Z]-+[0-9p-z]");
-        m = p.matcher("d__F-8");
-        assertTrue(m.matches());
-        m = p.matcher("c_a-q");
-        assertTrue(m.matches());
-        m = p.matcher("a__R-a");
-        assertFalse(m.matches());
-        m = p.matcher("r_____d-----5");
-        assertTrue(m.matches());
-
-        // Test range using unicode characters and unicode and hex escapes
-        p = Pattern.compile("[\\u1234-\\u2345]_+[a-z]-+[\u0001-\\x11]");
-        m = p.matcher("\u2000_q-\u0007");
-        assertTrue(m.matches());
-        m = p.matcher("\u1234_z-\u0001");
-        assertTrue(m.matches());
-        m = p.matcher("r_p-q");
-        assertFalse(m.matches());
-        m = p.matcher("\u2345_____d-----\n");
-        assertTrue(m.matches());
-
-// BEGIN android-removed
-// The "---" collides with ICU's "--" operator and is likely to be a user error
-// anyway, so we simply comment this one out.
-//        // Test ranges including the "-" character
-//        p = Pattern.compile("[\\*-/]_+[---]!+[--AP]");
-//        m = p.matcher("-_-!!A");
-//        assertTrue(m.matches());
-//        m = p.matcher("\u002b_-!!!-");
-//        assertTrue(m.matches());
-//        m = p.matcher("!_-!@");
-//        assertFalse(m.matches());
-//        m = p.matcher(",______-!!!!!!!P");
-//        assertTrue(m.matches());
-// END android-removed
-
-        // Test nested ranges
-        p = Pattern.compile("[pm[t]][a-z]+[[r]lp]");
-        m = p.matcher("prop");
-        assertTrue(m.matches());
-        m = p.matcher("tsar");
-        assertTrue(m.matches());
-        m = p.matcher("pong");
-        assertFalse(m.matches());
-        m = p.matcher("moor");
-        assertTrue(m.matches());
-
-        // Test character class intersection with &&
-        // TODO: figure out what x&&y or any class with a null intersection
-        // set (like [[a-c]&&[d-f]]) might mean. It doesn't mean "match
-        // nothing" and doesn't mean "match anything" so I'm stumped.
-        p = Pattern.compile("[[a-p]&&[g-z]]+-+[[a-z]&&q]-+[x&&[a-z]]-+");
-        m = p.matcher("h--q--x--");
-        assertTrue(m.matches());
-        m = p.matcher("hog--q-x-");
-        assertTrue(m.matches());
-        m = p.matcher("ape--q-x-");
-        assertFalse(m.matches());
-        m = p.matcher("mop--q-x----");
-        assertTrue(m.matches());
-
-        // Test error cases with &&
-// BEGIN android-removed
-// This is more of a bug, and ICU doesn't have this behavior.
-//            p = Pattern.compile("[&&[xyz]]");
-//            m = p.matcher("&");
-//            // System.out.println(m.matches());
-//            m = p.matcher("x");
-//            // System.out.println(m.matches());
-//            m = p.matcher("y");
-//            // System.out.println(m.matches());
-// END android-removed
-            p = Pattern.compile("[[xyz]&[axy]]");
-            m = p.matcher("x");
-            // System.out.println(m.matches());
-            m = p.matcher("z");
-            // System.out.println(m.matches());
-            m = p.matcher("&");
-            // System.out.println(m.matches());
-            p = Pattern.compile("[abc[123]&&[345]def]");
-            m = p.matcher("a");
-            // System.out.println(m.matches());
-
-// BEGIN android-removed
-// This is more of a bug, and ICU doesn't have this behavior.
-//            p = Pattern.compile("[[xyz]&&]");
-// END android-removed
-            p = Pattern.compile("[[abc]&]");
-
-        try {
-            p = Pattern.compile("[[abc]&&");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        p = Pattern.compile("[[abc]\\&&[xyz]]");
-
-        p = Pattern.compile("[[abc]&\\&[xyz]]");
-
-        // Test 3-way intersection
-        p = Pattern.compile("[[a-p]&&[g-z]&&[d-k]]");
-        m = p.matcher("g");
-        assertTrue(m.matches());
-        m = p.matcher("m");
-        assertFalse(m.matches());
-
-        // Test nested intersection
-        p = Pattern.compile("[[[a-p]&&[g-z]]&&[d-k]]");
-        m = p.matcher("g");
-        assertTrue(m.matches());
-        m = p.matcher("m");
-        assertFalse(m.matches());
-
-        // Test character class subtraction with && and ^
-        p = Pattern.compile("[[a-z]&&[^aeiou]][aeiou][[^xyz]&&[a-z]]");
-        m = p.matcher("pop");
-        assertTrue(m.matches());
-        m = p.matcher("tag");
-        assertTrue(m.matches());
-        m = p.matcher("eat");
-        assertFalse(m.matches());
-        m = p.matcher("tax");
-        assertFalse(m.matches());
-        m = p.matcher("zip");
-        assertTrue(m.matches());
-
-        // Test . (DOT), with and without DOTALL
-        // Note: DOT not allowed in character classes
-        p = Pattern.compile(".+/x.z");
-        m = p.matcher("!$/xyz");
-        assertTrue(m.matches());
-        m = p.matcher("%\n\r/x\nz");
-        assertFalse(m.matches());
-        p = Pattern.compile(".+/x.z", Pattern.DOTALL);
-        m = p.matcher("%\n\r/x\nz");
-        assertTrue(m.matches());
-
-        // Test \d (digit)
-        p = Pattern.compile("\\d+[a-z][\\dx]");
-        m = p.matcher("42a6");
-        assertTrue(m.matches());
-        m = p.matcher("21zx");
-        assertTrue(m.matches());
-        m = p.matcher("ab6");
-        assertFalse(m.matches());
-        m = p.matcher("56912f9");
-        assertTrue(m.matches());
-
-        // Test \D (not a digit)
-        p = Pattern.compile("\\D+[a-z]-[\\D3]");
-        m = p.matcher("za-p");
-        assertTrue(m.matches());
-        m = p.matcher("%!e-3");
-        assertTrue(m.matches());
-        m = p.matcher("9a-x");
-        assertFalse(m.matches());
-        m = p.matcher("\u1234pp\ny-3");
-        assertTrue(m.matches());
-
-        // Test \s (whitespace)
-        p = Pattern.compile("<[a-zA-Z]+\\s+[0-9]+[\\sx][^\\s]>");
-        m = p.matcher("<cat \t1\fx>");
-        assertTrue(m.matches());
-        m = p.matcher("<cat \t1\f >");
-        assertFalse(m.matches());
-        m = p
-                .matcher("xyz <foo\n\r22 5> <pp \t\n\f\r \u000b41x\u1234><pp \nx7\rc> zzz");
-        assertTrue(m.find());
-        assertTrue(m.find());
-        assertFalse(m.find());
-
-        // Test \S (not whitespace)
-        p = Pattern.compile("<[a-z] \\S[0-9][\\S\n]+[^\\S]221>");
-        m = p.matcher("<f $0**\n** 221>");
-        assertTrue(m.matches());
-        m = p.matcher("<x 441\t221>");
-        assertTrue(m.matches());
-        m = p.matcher("<z \t9\ng 221>");
-        assertFalse(m.matches());
-        m = p.matcher("<z 60\ngg\u1234\f221>");
-        assertTrue(m.matches());
-        p = Pattern.compile("<[a-z] \\S[0-9][\\S\n]+[^\\S]221[\\S&&[^abc]]>");
-        m = p.matcher("<f $0**\n** 221x>");
-        assertTrue(m.matches());
-        m = p.matcher("<x 441\t221z>");
-        assertTrue(m.matches());
-        m = p.matcher("<x 441\t221 >");
-        assertFalse(m.matches());
-        m = p.matcher("<x 441\t221c>");
-        assertFalse(m.matches());
-        m = p.matcher("<z \t9\ng 221x>");
-        assertFalse(m.matches());
-        m = p.matcher("<z 60\ngg\u1234\f221\u0001>");
-        assertTrue(m.matches());
-
-        // Test \w (ascii word)
-        p = Pattern.compile("<\\w+\\s[0-9]+;[^\\w]\\w+/[\\w$]+;");
-        m = p.matcher("<f1 99;!foo5/a$7;");
-        assertTrue(m.matches());
-        m = p.matcher("<f$ 99;!foo5/a$7;");
-        assertFalse(m.matches());
-        m = p
-                .matcher("<abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_0123456789 99;!foo5/a$7;");
-        assertTrue(m.matches());
-
-        // Test \W (not an ascii word)
-        p = Pattern.compile("<\\W\\w+\\s[0-9]+;[\\W_][^\\W]+\\s[0-9]+;");
-        m = p.matcher("<$foo3\n99;_bar\t0;");
-        assertTrue(m.matches());
-        m = p.matcher("<hh 99;_g 0;");
-        assertFalse(m.matches());
-        m = p.matcher("<*xx\t00;^zz\f11;");
-        assertTrue(m.matches());
-
-        // Test x|y pattern
-        // TODO
-    }
-    public void testPOSIXGroups() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        // Test POSIX groups using \p and \P (in the group and not in the group)
-        // Groups are Lower, Upper, ASCII, Alpha, Digit, XDigit, Alnum, Punct,
-        // Graph, Print, Blank, Space, Cntrl
-        // Test \p{Lower}
-        /*
-         * FIXME: Requires complex range processing
-         * p = Pattern.compile("<\\p{Lower}\\d\\P{Lower}:[\\p{Lower}Z]\\s[^\\P{Lower}]>");
-         * m = p.matcher("<a4P:g x>"); assertTrue(m.matches()); m =
-         * p.matcher("<p4%:Z\tq>"); assertTrue(m.matches()); m =
-         * p.matcher("<A6#:e e>"); assertFalse(m.matches());
-         */
-        p = Pattern.compile("\\p{Lower}+");
-        m = p.matcher("abcdefghijklmnopqrstuvwxyz");
-        assertTrue(m.matches());
-
-        // Invalid uses of \p{Lower}
-        try {
-            p = Pattern.compile("\\p");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{Lower");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{Lower;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        // Test \p{Upper}
-        /*
-         * FIXME: Requires complex range processing
-         * p = Pattern.compile("<\\p{Upper}\\d\\P{Upper}:[\\p{Upper}z]\\s[^\\P{Upper}]>");
-         * m = p.matcher("<A4p:G X>"); assertTrue(m.matches()); m =
-         * p.matcher("<P4%:z\tQ>"); assertTrue(m.matches()); m =
-         * p.matcher("<a6#:E E>"); assertFalse(m.matches());
-         */
-        p = Pattern.compile("\\p{Upper}+");
-        m = p.matcher("ABCDEFGHIJKLMNOPQRSTUVWXYZ");
-        assertTrue(m.matches());
-
-        // Invalid uses of \p{Upper}
-        try {
-            p = Pattern.compile("\\p{Upper");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{Upper;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        // Test \p{ASCII}
-        /*
-         * FIXME: Requires complex range processing p = Pattern.compile("<\\p{ASCII}\\d\\P{ASCII}:[\\p{ASCII}\u1234]\\s[^\\P{ASCII}]>");
-         * m = p.matcher("<A4\u0080:G X>"); assertTrue(m.matches()); m =
-         * p.matcher("<P4\u00ff:\u1234\t\n>"); assertTrue(m.matches()); m =
-         * p.matcher("<\u00846#:E E>"); assertFalse(m.matches())
-         */
-        int i;
-        p = Pattern.compile("\\p{ASCII}");
-        for (i = 0; i < 0x80; i++) {
-            m = p.matcher(Character.toString((char) i));
-            assertTrue(m.matches());
-        }
-        for (; i < 0xff; i++) {
-            m = p.matcher(Character.toString((char) i));
-            assertFalse(m.matches());
-        }
-
-        // Invalid uses of \p{ASCII}
-        try {
-            p = Pattern.compile("\\p{ASCII");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        try {
-            p = Pattern.compile("\\p{ASCII;");
-            fail("PatternSyntaxException expected");
-        } catch (PatternSyntaxException e) {
-        }
-
-        // Test \p{Alpha}
-        // TODO
-
-        // Test \p{Digit}
-        // TODO
-
-        // Test \p{XDigit}
-        // TODO
-
-        // Test \p{Alnum}
-        // TODO
-
-        // Test \p{Punct}
-        // TODO
-
-        // Test \p{Graph}
-        // TODO
-
-        // Test \p{Print}
-        // TODO
-
-        // Test \p{Blank}
-        // TODO
-
-        // Test \p{Space}
-        // TODO
-
-        // Test \p{Cntrl}
-        // TODO
-    }
     public void testUnicodeCategories() throws PatternSyntaxException {
         // Test Unicode categories using \p and \P
         // One letter codes: L, M, N, P, S, Z, C
@@ -906,75 +128,6 @@ public class Pattern2Test extends TestCase {
         }
     }
 
-    public void testUnicodeBlocks() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-        int i, j;
-
-        // Test Unicode blocks using \p and \P
-        // FIXME:
-        // Note that LatinExtended-B and ArabicPresentations-B are unrecognized
-        // by the reference JDK.
-        for (i = 0; i < UBlocks.length; i++) {
-            /*
-             * p = Pattern.compile("\\p{"+UBlocks[i].name+"}");
-             *
-             * if (UBlocks[i].low > 0) { m =
-             * p.matcher(Character.toString((char)(UBlocks[i].low-1)));
-             * assertFalse(m.matches()); } for (j=UBlocks[i].low; j <=
-             * UBlocks[i].high; j++) { m =
-             * p.matcher(Character.toString((char)j));
-             * assertTrue(m.matches()); } if (UBlocks[i].high < 0xFFFF) { m =
-             * p.matcher(Character.toString((char)(UBlocks[i].high+1)));
-             * assertFalse(m.matches()); }
-             *
-             * p = Pattern.compile("\\P{"+UBlocks[i].name+"}");
-             *
-             * if (UBlocks[i].low > 0) { m =
-             * p.matcher(Character.toString((char)(UBlocks[i].low-1)));
-             * assertTrue(m.matches()); } for (j=UBlocks[i].low; j <
-             * UBlocks[i].high; j++) { m =
-             * p.matcher(Character.toString((char)j));
-             * assertFalse(m.matches()); } if (UBlocks[i].high < 0xFFFF) { m =
-             * p.matcher(Character.toString((char)(UBlocks[i].high+1)));
-             * assertTrue(m.matches()); }
-             */
-
-            p = Pattern.compile("\\p{In" + UBlocks[i].name + "}");
-// BEGIN android-changed
-// Added the name of the block under test to the assertion to get more output.
-
-            if (UBlocks[i].low > 0) {
-                m = p.matcher(Character.toString((char) (UBlocks[i].low - 1)));
-                assertFalse(UBlocks[i].name, m.matches());
-            }
-            for (j = UBlocks[i].low; j <= UBlocks[i].high; j++) {
-                m = p.matcher(Character.toString((char) j));
-                assertTrue(UBlocks[i].name, m.matches());
-            }
-            if (UBlocks[i].high < 0xFFFF) {
-                m = p.matcher(Character.toString((char) (UBlocks[i].high + 1)));
-                assertFalse(UBlocks[i].name, m.matches());
-            }
-
-            p = Pattern.compile("\\P{In" + UBlocks[i].name + "}");
-
-            if (UBlocks[i].low > 0) {
-                m = p.matcher(Character.toString((char) (UBlocks[i].low - 1)));
-                assertTrue(UBlocks[i].name, m.matches());
-            }
-            for (j = UBlocks[i].low; j < UBlocks[i].high; j++) {
-                m = p.matcher(Character.toString((char) j));
-                assertFalse(UBlocks[i].name, m.matches());
-            }
-            if (UBlocks[i].high < 0xFFFF) {
-                m = p.matcher(Character.toString((char) (UBlocks[i].high + 1)));
-                assertTrue(UBlocks[i].name, m.matches());
-            }
-
-// END android-changed
-        }
-    }
     public void testCapturingGroups() throws PatternSyntaxException {
         Pattern p;
         Matcher m;
@@ -1142,436 +295,4 @@ public class Pattern2Test extends TestCase {
         // Test negative lookbehind using (?<!...)
         // TODO
     }
-    public void testMisc() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-
-        // Test (?>...)
-        // TODO
-
-        // Test (?onflags-offflags)
-        // Valid flags are i,m,d,s,u,x
-        // TODO
-
-        // Test (?onflags-offflags:...)
-        // TODO
-
-        // Test \Q, \E
-        p = Pattern.compile("[a-z]+;\\Q[a-z]+;\\Q(foo.*);\\E[0-9]+");
-        m = p.matcher("abc;[a-z]+;\\Q(foo.*);411");
-        assertTrue(m.matches());
-        m = p.matcher("abc;def;foo42;555");
-        assertFalse(m.matches());
-        m = p.matcher("abc;\\Qdef;\\Qfoo99;\\E123");
-        assertFalse(m.matches());
-
-        p = Pattern.compile("[a-z]+;(foo[0-9]-\\Q(...)\\E);[0-9]+");
-        m = p.matcher("abc;foo5-(...);123");
-        assertTrue(m.matches());
-        assertEquals("foo5-(...)", m.group(1));
-        m = p.matcher("abc;foo9-(xxx);789");
-        assertFalse(m.matches());
-
-        p = Pattern.compile("[a-z]+;(bar[0-9]-[a-z\\Q$-\\E]+);[0-9]+");
-        m = p.matcher("abc;bar0-def$-;123");
-        assertTrue(m.matches());
-
-        // FIXME:
-        // This should work the same as the pattern above but fails with the
-        // the reference JDK
-        p = Pattern.compile("[a-z]+;(bar[0-9]-[a-z\\Q-$\\E]+);[0-9]+");
-        m = p.matcher("abc;bar0-def$-;123");
-        // assertTrue(m.matches());
-
-        // FIXME:
-        // This should work too .. it looks as if just about anything that
-        // has more
-        // than one character between \Q and \E is broken in the the reference JDK
-        p = Pattern.compile("[a-z]+;(bar[0-9]-[a-z\\Q[0-9]\\E]+);[0-9]+");
-        m = p.matcher("abc;bar0-def[99]-]0x[;123");
-        // assertTrue(m.matches());
-
-        // This is the same as above but with explicit escapes .. and this
-        // does work
-        // on the the reference JDK
-        p = Pattern.compile("[a-z]+;(bar[0-9]-[a-z\\[0\\-9\\]]+);[0-9]+");
-        m = p.matcher("abc;bar0-def[99]-]0x[;123");
-        assertTrue(m.matches());
-
-        // Test #<comment text>
-        // TODO
-    }
-    public void testCompile1() throws PatternSyntaxException {
-        Pattern pattern = Pattern
-                .compile("[0-9A-Za-z][0-9A-Za-z\\x2e\\x3a\\x2d\\x5f]*");
-        String name = "iso-8859-1";
-        assertTrue(pattern.matcher(name).matches());
-    }
-    public void testCompile2() throws PatternSyntaxException {
-        String findString = "\\Qimport\\E";
-
-        Pattern pattern = Pattern.compile(findString, 0);
-        Matcher matcher = pattern.matcher(new String(
-                "import a.A;\n\n import b.B;\nclass C {}"));
-
-        assertTrue(matcher.find(0));
-    }
-    public void testCompile3() throws PatternSyntaxException {
-        Pattern p;
-        Matcher m;
-        p = Pattern.compile("a$");
-        m = p.matcher("a\n");
-        assertTrue(m.find());
-        assertEquals("a", m.group());
-        assertFalse(m.find());
-
-        p = Pattern.compile("(a$)");
-        m = p.matcher("a\n");
-        assertTrue(m.find());
-        assertEquals("a", m.group());
-        assertEquals("a", m.group(1));
-        assertFalse(m.find());
-
-        p = Pattern.compile("^.*$", Pattern.MULTILINE);
-
-        m = p.matcher("a\n");
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("a", m.group());
-        assertFalse(m.find());
-
-        m = p.matcher("a\nb\n");
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("a", m.group());
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("b", m.group());
-        assertFalse(m.find());
-
-        m = p.matcher("a\nb");
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("a", m.group());
-        assertTrue(m.find());
-        assertEquals("b", m.group());
-        assertFalse(m.find());
-
-        m = p.matcher("\naa\r\nbb\rcc\n\n");
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertTrue(m.group().equals(""));
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("aa", m.group());
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("bb", m.group());
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertEquals("cc", m.group());
-        assertTrue(m.find());
-        // System.out.println("["+m.group()+"]");
-        assertTrue(m.group().equals(""));
-        assertFalse(m.find());
-
-        m = p.matcher("a");
-        assertTrue(m.find());
-        assertEquals("a", m.group());
-        assertFalse(m.find());
-
-// BEGIN android-removed
-// Makes no sense to duplicate this weird behavior
-//        m = p.matcher("");
-//        // FIXME: This matches the reference behaviour but is
-//        // inconsistent with matching "a" - ie. the end of the
-//        // target string should match against $ always but this
-//        // appears to work with the null string only when not in
-//        // multiline mode (see below)
-//        assertFalse(m.find());
-// END android-removed
-
-        p = Pattern.compile("^.*$");
-        m = p.matcher("");
-        assertTrue(m.find());
-        assertTrue(m.group().equals(""));
-        assertFalse(m.find());
-    }
-    public void testCompile4() throws PatternSyntaxException {
-        String findString = "\\Qpublic\\E";
-        StringBuffer text = new StringBuffer("    public class Class {\n"
-                + "    public class Class {");
-
-        Pattern pattern = Pattern.compile(findString, 0);
-        Matcher matcher = pattern.matcher(text);
-
-        boolean found = matcher.find();
-        assertTrue(found);
-        assertEquals(4, matcher.start());
-        if (found) {
-            // modify text
-            text.delete(0, text.length());
-            text.append("Text have been changed.");
-            matcher.reset(text);
-        }
-
-        found = matcher.find();
-        assertFalse(found);
-    }
-    public void testCompile5() throws PatternSyntaxException {
-        Pattern p = Pattern.compile("^[0-9]");
-        String s[] = p.split("12", -1);
-        assertEquals("", s[0]);
-        assertEquals("2", s[1]);
-        assertEquals(2, s.length);
-    }
-
-    //      public void testCompile6() {
-    //        String regex = "[\\p{L}[\\p{Mn}[\\p{Pc}[\\p{Nd}[\\p{Nl}[\\p{Sc}]]]]]]+";
-    //        String regex = "[\\p{L}\\p{Mn}\\p{Pc}\\p{Nd}\\p{Nl}\\p{Sc}]+";
-    //        try {
-    //            Pattern pattern = Pattern.compile(regex, Pattern.MULTILINE);
-    //            assertTrue(true);
-    //        } catch (PatternSyntaxException e) {
-    //            System.out.println(e.getMessage());
-    //            assertTrue(false);
-    //        }
-    //    }
-
-    private static class UBInfo {
-        public UBInfo(int low, int high, String name) {
-            this.name = name;
-            this.low = low;
-            this.high = high;
-        }
-
-        public String name;
-
-        public int low, high;
-    }
-
-    // A table representing the unicode categories
-    //private static UBInfo[] UCategories = {
-    // Lu
-    // Ll
-    // Lt
-    // Lm
-    // Lo
-    // Mn
-    // Mc
-    // Me
-    // Nd
-    // Nl
-    // No
-    // Pc
-    // Pd
-    // Ps
-    // Pe
-    // Pi
-    // Pf
-    // Po
-    // Sm
-    // Sc
-    // Sk
-    // So
-    // Zs
-    // Zl
-    // Zp
-    // Cc
-    // Cf
-    // Cs
-    // Co
-    // Cn
-    //};
-
-    // A table representing the unicode character blocks
-    private static UBInfo[] UBlocks = {
-    /* 0000; 007F; Basic Latin */
-    new UBInfo(0x0000, 0x007F, "BasicLatin"), // Character.UnicodeBlock.BASIC_LATIN
-            /* 0080; 00FF; Latin-1 Supplement */
-            new UBInfo(0x0080, 0x00FF, "Latin-1Supplement"), // Character.UnicodeBlock.LATIN_1_SUPPLEMENT
-            /* 0100; 017F; Latin Extended-A */
-            new UBInfo(0x0100, 0x017F, "LatinExtended-A"), // Character.UnicodeBlock.LATIN_EXTENDED_A
-            /* 0180; 024F; Latin Extended-B */
-            // new UBInfo (0x0180,0x024F,"InLatinExtended-B"), //
-            // Character.UnicodeBlock.LATIN_EXTENDED_B
-            /* 0250; 02AF; IPA Extensions */
-            new UBInfo(0x0250, 0x02AF, "IPAExtensions"), // Character.UnicodeBlock.IPA_EXTENSIONS
-            /* 02B0; 02FF; Spacing Modifier Letters */
-            new UBInfo(0x02B0, 0x02FF, "SpacingModifierLetters"), // Character.UnicodeBlock.SPACING_MODIFIER_LETTERS
-            /* 0300; 036F; Combining Diacritical Marks */
-            new UBInfo(0x0300, 0x036F, "CombiningDiacriticalMarks"), // Character.UnicodeBlock.COMBINING_DIACRITICAL_MARKS
-            /* 0370; 03FF; Greek */
-            new UBInfo(0x0370, 0x03FF, "Greek"), // Character.UnicodeBlock.GREEK
-            /* 0400; 04FF; Cyrillic */
-            new UBInfo(0x0400, 0x04FF, "Cyrillic"), // Character.UnicodeBlock.CYRILLIC
-            /* 0530; 058F; Armenian */
-            new UBInfo(0x0530, 0x058F, "Armenian"), // Character.UnicodeBlock.ARMENIAN
-            /* 0590; 05FF; Hebrew */
-            new UBInfo(0x0590, 0x05FF, "Hebrew"), // Character.UnicodeBlock.HEBREW
-            /* 0600; 06FF; Arabic */
-            new UBInfo(0x0600, 0x06FF, "Arabic"), // Character.UnicodeBlock.ARABIC
-            /* 0700; 074F; Syriac */
-            new UBInfo(0x0700, 0x074F, "Syriac"), // Character.UnicodeBlock.SYRIAC
-            /* 0780; 07BF; Thaana */
-            new UBInfo(0x0780, 0x07BF, "Thaana"), // Character.UnicodeBlock.THAANA
-            /* 0900; 097F; Devanagari */
-            new UBInfo(0x0900, 0x097F, "Devanagari"), // Character.UnicodeBlock.DEVANAGARI
-            /* 0980; 09FF; Bengali */
-            new UBInfo(0x0980, 0x09FF, "Bengali"), // Character.UnicodeBlock.BENGALI
-            /* 0A00; 0A7F; Gurmukhi */
-            new UBInfo(0x0A00, 0x0A7F, "Gurmukhi"), // Character.UnicodeBlock.GURMUKHI
-            /* 0A80; 0AFF; Gujarati */
-            new UBInfo(0x0A80, 0x0AFF, "Gujarati"), // Character.UnicodeBlock.GUJARATI
-            /* 0B00; 0B7F; Oriya */
-            new UBInfo(0x0B00, 0x0B7F, "Oriya"), // Character.UnicodeBlock.ORIYA
-            /* 0B80; 0BFF; Tamil */
-            new UBInfo(0x0B80, 0x0BFF, "Tamil"), // Character.UnicodeBlock.TAMIL
-            /* 0C00; 0C7F; Telugu */
-            new UBInfo(0x0C00, 0x0C7F, "Telugu"), // Character.UnicodeBlock.TELUGU
-            /* 0C80; 0CFF; Kannada */
-            new UBInfo(0x0C80, 0x0CFF, "Kannada"), // Character.UnicodeBlock.KANNADA
-            /* 0D00; 0D7F; Malayalam */
-            new UBInfo(0x0D00, 0x0D7F, "Malayalam"), // Character.UnicodeBlock.MALAYALAM
-            /* 0D80; 0DFF; Sinhala */
-            new UBInfo(0x0D80, 0x0DFF, "Sinhala"), // Character.UnicodeBlock.SINHALA
-            /* 0E00; 0E7F; Thai */
-            new UBInfo(0x0E00, 0x0E7F, "Thai"), // Character.UnicodeBlock.THAI
-            /* 0E80; 0EFF; Lao */
-            new UBInfo(0x0E80, 0x0EFF, "Lao"), // Character.UnicodeBlock.LAO
-            /* 0F00; 0FFF; Tibetan */
-            new UBInfo(0x0F00, 0x0FFF, "Tibetan"), // Character.UnicodeBlock.TIBETAN
-            /* 1000; 109F; Myanmar */
-            new UBInfo(0x1000, 0x109F, "Myanmar"), // Character.UnicodeBlock.MYANMAR
-            /* 10A0; 10FF; Georgian */
-            new UBInfo(0x10A0, 0x10FF, "Georgian"), // Character.UnicodeBlock.GEORGIAN
-            /* 1100; 11FF; Hangul Jamo */
-            new UBInfo(0x1100, 0x11FF, "HangulJamo"), // Character.UnicodeBlock.HANGUL_JAMO
-            /* 1200; 137F; Ethiopic */
-            new UBInfo(0x1200, 0x137F, "Ethiopic"), // Character.UnicodeBlock.ETHIOPIC
-            /* 13A0; 13FF; Cherokee */
-            new UBInfo(0x13A0, 0x13FF, "Cherokee"), // Character.UnicodeBlock.CHEROKEE
-            /* 1400; 167F; Unified Canadian Aboriginal Syllabics */
-            new UBInfo(0x1400, 0x167F, "UnifiedCanadianAboriginalSyllabics"), // Character.UnicodeBlock.UNIFIED_CANADIAN_ABORIGINAL_SYLLABICS
-            /* 1680; 169F; Ogham */
-            new UBInfo(0x1680, 0x169F, "Ogham"), // Character.UnicodeBlock.OGHAM
-            /* 16A0; 16FF; Runic */
-            new UBInfo(0x16A0, 0x16FF, "Runic"), // Character.UnicodeBlock.RUNIC
-            /* 1780; 17FF; Khmer */
-            new UBInfo(0x1780, 0x17FF, "Khmer"), // Character.UnicodeBlock.KHMER
-            /* 1800; 18AF; Mongolian */
-            new UBInfo(0x1800, 0x18AF, "Mongolian"), // Character.UnicodeBlock.MONGOLIAN
-            /* 1E00; 1EFF; Latin Extended Additional */
-            new UBInfo(0x1E00, 0x1EFF, "LatinExtendedAdditional"), // Character.UnicodeBlock.LATIN_EXTENDED_ADDITIONAL
-            /* 1F00; 1FFF; Greek Extended */
-            new UBInfo(0x1F00, 0x1FFF, "GreekExtended"), // Character.UnicodeBlock.GREEK_EXTENDED
-            /* 2000; 206F; General Punctuation */
-            new UBInfo(0x2000, 0x206F, "GeneralPunctuation"), // Character.UnicodeBlock.GENERAL_PUNCTUATION
-            /* 2070; 209F; Superscripts and Subscripts */
-            new UBInfo(0x2070, 0x209F, "SuperscriptsandSubscripts"), // Character.UnicodeBlock.SUPERSCRIPTS_AND_SUBSCRIPTS
-            /* 20A0; 20CF; Currency Symbols */
-            new UBInfo(0x20A0, 0x20CF, "CurrencySymbols"), // Character.UnicodeBlock.CURRENCY_SYMBOLS
-            /* 20D0; 20FF; Combining Marks for Symbols */
-            new UBInfo(0x20D0, 0x20FF, "CombiningMarksforSymbols"), // Character.UnicodeBlock.COMBINING_MARKS_FOR_SYMBOLS
-            /* 2100; 214F; Letterlike Symbols */
-            new UBInfo(0x2100, 0x214F, "LetterlikeSymbols"), // Character.UnicodeBlock.LETTERLIKE_SYMBOLS
-            /* 2150; 218F; Number Forms */
-            new UBInfo(0x2150, 0x218F, "NumberForms"), // Character.UnicodeBlock.NUMBER_FORMS
-            /* 2190; 21FF; Arrows */
-            new UBInfo(0x2190, 0x21FF, "Arrows"), // Character.UnicodeBlock.ARROWS
-            /* 2200; 22FF; Mathematical Operators */
-            new UBInfo(0x2200, 0x22FF, "MathematicalOperators"), // Character.UnicodeBlock.MATHEMATICAL_OPERATORS
-            /* 2300; 23FF; Miscellaneous Technical */
-            new UBInfo(0x2300, 0x23FF, "MiscellaneousTechnical"), // Character.UnicodeBlock.MISCELLANEOUS_TECHNICAL
-            /* 2400; 243F; Control Pictures */
-            new UBInfo(0x2400, 0x243F, "ControlPictures"), // Character.UnicodeBlock.CONTROL_PICTURES
-            /* 2440; 245F; Optical Character Recognition */
-            new UBInfo(0x2440, 0x245F, "OpticalCharacterRecognition"), // Character.UnicodeBlock.OPTICAL_CHARACTER_RECOGNITION
-            /* 2460; 24FF; Enclosed Alphanumerics */
-            new UBInfo(0x2460, 0x24FF, "EnclosedAlphanumerics"), // Character.UnicodeBlock.ENCLOSED_ALPHANUMERICS
-            /* 2500; 257F; Box Drawing */
-            new UBInfo(0x2500, 0x257F, "BoxDrawing"), // Character.UnicodeBlock.BOX_DRAWING
-            /* 2580; 259F; Block Elements */
-            new UBInfo(0x2580, 0x259F, "BlockElements"), // Character.UnicodeBlock.BLOCK_ELEMENTS
-            /* 25A0; 25FF; Geometric Shapes */
-            new UBInfo(0x25A0, 0x25FF, "GeometricShapes"), // Character.UnicodeBlock.GEOMETRIC_SHAPES
-            /* 2600; 26FF; Miscellaneous Symbols */
-            new UBInfo(0x2600, 0x26FF, "MiscellaneousSymbols"), // Character.UnicodeBlock.MISCELLANEOUS_SYMBOLS
-            /* 2700; 27BF; Dingbats */
-            new UBInfo(0x2700, 0x27BF, "Dingbats"), // Character.UnicodeBlock.DINGBATS
-            /* 2800; 28FF; Braille Patterns */
-            new UBInfo(0x2800, 0x28FF, "BraillePatterns"), // Character.UnicodeBlock.BRAILLE_PATTERNS
-            /* 2E80; 2EFF; CJK Radicals Supplement */
-            new UBInfo(0x2E80, 0x2EFF, "CJKRadicalsSupplement"), // Character.UnicodeBlock.CJK_RADICALS_SUPPLEMENT
-            /* 2F00; 2FDF; Kangxi Radicals */
-            new UBInfo(0x2F00, 0x2FDF, "KangxiRadicals"), // Character.UnicodeBlock.KANGXI_RADICALS
-            /* 2FF0; 2FFF; Ideographic Description Characters */
-            new UBInfo(0x2FF0, 0x2FFF, "IdeographicDescriptionCharacters"), // Character.UnicodeBlock.IDEOGRAPHIC_DESCRIPTION_CHARACTERS
-            /* 3000; 303F; CJK Symbols and Punctuation */
-            new UBInfo(0x3000, 0x303F, "CJKSymbolsandPunctuation"), // Character.UnicodeBlock.CJK_SYMBOLS_AND_PUNCTUATION
-            /* 3040; 309F; Hiragana */
-            new UBInfo(0x3040, 0x309F, "Hiragana"), // Character.UnicodeBlock.HIRAGANA
-            /* 30A0; 30FF; Katakana */
-            new UBInfo(0x30A0, 0x30FF, "Katakana"), // Character.UnicodeBlock.KATAKANA
-            /* 3100; 312F; Bopomofo */
-            new UBInfo(0x3100, 0x312F, "Bopomofo"), // Character.UnicodeBlock.BOPOMOFO
-            /* 3130; 318F; Hangul Compatibility Jamo */
-            new UBInfo(0x3130, 0x318F, "HangulCompatibilityJamo"), // Character.UnicodeBlock.HANGUL_COMPATIBILITY_JAMO
-            /* 3190; 319F; Kanbun */
-            new UBInfo(0x3190, 0x319F, "Kanbun"), // Character.UnicodeBlock.KANBUN
-            /* 31A0; 31BF; Bopomofo Extended */
-            new UBInfo(0x31A0, 0x31BF, "BopomofoExtended"), // Character.UnicodeBlock.BOPOMOFO_EXTENDED
-            /* 3200; 32FF; Enclosed CJK Letters and Months */
-            new UBInfo(0x3200, 0x32FF, "EnclosedCJKLettersandMonths"), // Character.UnicodeBlock.ENCLOSED_CJK_LETTERS_AND_MONTHS
-            /* 3300; 33FF; CJK Compatibility */
-            new UBInfo(0x3300, 0x33FF, "CJKCompatibility"), // Character.UnicodeBlock.CJK_COMPATIBILITY
-            /* 3400; 4DB5; CJK Unified Ideographs Extension A */
-// BEGIN android-changed
-// Modified this to reflect current Unicode tables (or maybe it was a typo)
-            new UBInfo(0x3400, 0x4DBF, "CJKUnifiedIdeographsExtensionA"), // Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A
-// END android-changed
-            /* 4E00; 9FFF; CJK Unified Ideographs */
-            new UBInfo(0x4E00, 0x9FFF, "CJKUnifiedIdeographs"), // Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS
-            /* A000; A48F; Yi Syllables */
-            new UBInfo(0xA000, 0xA48F, "YiSyllables"), // Character.UnicodeBlock.YI_SYLLABLES
-            /* A490; A4CF; Yi Radicals */
-            new UBInfo(0xA490, 0xA4CF, "YiRadicals"), // Character.UnicodeBlock.YI_RADICALS
-            /* AC00; D7A3; Hangul Syllables */
-// BEGIN android-changed
-// Modified this to reflect current Unicode tables (or maybe it was a typo)
-            new UBInfo(0xAC00, 0xD7AF, "HangulSyllables"), // Character.UnicodeBlock.HANGUL_SYLLABLES
-// END android-changed
-            /* D800; DB7F; High Surrogates */
-            /* DB80; DBFF; High Private Use Surrogates */
-            /* DC00; DFFF; Low Surrogates */
-            /* E000; F8FF; Private Use */
-            /* F900; FAFF; CJK Compatibility Ideographs */
-            new UBInfo(0xF900, 0xFAFF, "CJKCompatibilityIdeographs"), // Character.UnicodeBlock.CJK_COMPATIBILITY_IDEOGRAPHS
-            /* FB00; FB4F; Alphabetic Presentation Forms */
-            new UBInfo(0xFB00, 0xFB4F, "AlphabeticPresentationForms"), // Character.UnicodeBlock.ALPHABETIC_PRESENTATION_FORMS
-            /* FB50; FDFF; Arabic Presentation Forms-A */
-            new UBInfo(0xFB50, 0xFDFF, "ArabicPresentationForms-A"), // Character.UnicodeBlock.ARABIC_PRESENTATION_FORMS_A
-            /* FE20; FE2F; Combining Half Marks */
-            new UBInfo(0xFE20, 0xFE2F, "CombiningHalfMarks"), // Character.UnicodeBlock.COMBINING_HALF_MARKS
-            /* FE30; FE4F; CJK Compatibility Forms */
-            new UBInfo(0xFE30, 0xFE4F, "CJKCompatibilityForms"), // Character.UnicodeBlock.CJK_COMPATIBILITY_FORMS
-            /* FE50; FE6F; Small Form Variants */
-            new UBInfo(0xFE50, 0xFE6F, "SmallFormVariants"), // Character.UnicodeBlock.SMALL_FORM_VARIANTS
-            /* FE70; FEFE; Arabic Presentation Forms-B */
-            // new UBInfo (0xFE70,0xFEFE,"InArabicPresentationForms-B"), //
-            // Character.UnicodeBlock.ARABIC_PRESENTATION_FORMS_B
-            /* FEFF; FEFF; Specials */
-// BEGIN android-changed
-// Modified this to reflect current Unicode tables (or maybe it was a typo)
-// FEFF is actually still Arabic Presentation Forms B
-//            new UBInfo(0xFEFF, 0xFEFF, "Specials"), // Character.UnicodeBlock.SPECIALS
-// END android-changed
-            /* FF00; FFEF; Halfwidth and Fullwidth Forms */
-            new UBInfo(0xFF00, 0xFFEF, "HalfwidthandFullwidthForms"), // Character.UnicodeBlock.HALFWIDTH_AND_FULLWIDTH_FORMS
-            /* FFF0; FFFD; Specials */
-            // BEGIN android-changed
-// Modified this to reflect current Unicode tables (or maybe it was a typo)
-            new UBInfo(0xFFF0, 0xFFFF, "Specials") // Character.UnicodeBlock.SPECIALS
-// END android-changed
-    };
 }
diff --git a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternErrorTest.java b/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternErrorTest.java
deleted file mode 100644
index fd2c3ae..0000000
--- a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternErrorTest.java
+++ /dev/null
@@ -1,83 +0,0 @@
-/* Licensed to the Apache Software Foundation (ASF) under one or more
- * contributor license agreements.  See the NOTICE file distributed with
- * this work for additional information regarding copyright ownership.
- * The ASF licenses this file to You under the Apache License, Version 2.0
- * (the "License"); you may not use this file except in compliance with
- * the License.  You may obtain a copy of the License at
- *
- *     http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package org.apache.harmony.regex.tests.java.util.regex;
-
-import java.util.regex.Pattern;
-import junit.framework.TestCase;
-
-/**
- * Test boundary and error conditions in java.util.regex.Pattern
- *
- */
-public class PatternErrorTest extends TestCase {
-    public void testCompileErrors() throws Exception {
-        // null regex string - should get NullPointerException
-        try {
-            Pattern.compile(null);
-            fail("NullPointerException expected");
-        } catch (NullPointerException e) {
-        }
-
-
-        // empty regex string - no exception should be thrown
-        Pattern.compile("");
-
-        // note: invalid regex syntax checked in PatternSyntaxExceptionTest
-
-        // flags = 0 should raise no exception
-        int flags = 0;
-        Pattern.compile("foo", flags);
-
-        // check that all valid flags accepted without exception
-        flags |= Pattern.UNIX_LINES;
-        flags |= Pattern.CASE_INSENSITIVE;
-        flags |= Pattern.MULTILINE;
-        // BEGIN android-changed
-        // We don't support that flag.
-        // flags |= Pattern.CANON_EQ;
-        // END android-changed
-        flags |= Pattern.COMMENTS;
-        flags |= Pattern.DOTALL;
-        flags |= Pattern.UNICODE_CASE;
-        Pattern.compile("foo", flags);
-
-        // add invalid flags - should get IllegalArgumentException
-        /*
-         * TODO: Inconsistency between the reference JDK behaviour and spec - exception is
-         * not thrown
-         */
-        /*
-         * Valid test is:
-         * flags |= 0xFFFFFFFF;
-         * try {
-         *   Pattern.compile("foo",flags);
-         * } catch (IllegalArgumentException e) {
-         *   // This is the expected exception
-         * } catch (Exception e) {
-         *   fail();
-         * }
-         */
-
-        /* Workaround test is: */
-        // BEGIN android-changed
-        // We don't support that flag.
-        flags |= ~Pattern.CANON_EQ;
-        // END android-changed
-        // No exception expected to match incorrect the reference behaviour
-        Pattern.compile("foo", flags);
-    }
-}
diff --git a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternTest.java b/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternTest.java
index a81d294..7dfcabf 100644
--- a/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternTest.java
+++ b/luni/src/test/java/org/apache/harmony/regex/tests/java/util/regex/PatternTest.java
@@ -1020,24 +1020,6 @@ public class PatternTest extends TestCase {
         assertTrue(mat.matches());
     }
 
-    public void testCompileNonCaptGroup() {
-        boolean isCompiled = false;
-
-        try {
-// BEGIN android-change
-// We don't have canonical equivalence.
-            Pattern pat = Pattern.compile("(?:)");
-            pat = Pattern.compile("(?:)", Pattern.DOTALL);
-            pat = Pattern.compile("(?:)", Pattern.CASE_INSENSITIVE);
-            pat = Pattern.compile("(?:)", Pattern.COMMENTS | Pattern.UNIX_LINES);
-// END android-change
-            isCompiled = true;
-        } catch (PatternSyntaxException e) {
-            System.out.println(e);
-        }
-        assertTrue(isCompiled);
-    }
-
     public void testEmbeddedFlags() {
         String baseString = "(?i)((?s)a)";
         String testString = "A";