Regenerate the TopLevelDomain from iana.org website.

This is to capture recently added top level domains.
	modified:   common/java/com/android/common/Patterns.java

	modified:   common/tests/src/com/android/common/PatternsTest.java

2 files changed, 100 insertions, 74 deletions

diff --git a/common/java/com/android/common/Patterns.java b/common/java/com/android/common/Patterns.java
index 24a18c0..71c3a5e 100644
--- a/common/java/com/android/common/Patterns.java
+++ b/common/java/com/android/common/Patterns.java
@@ -25,87 +25,87 @@ import java.util.regex.Pattern;
 public class Patterns {
     /**
      *  Regular expression pattern to match all IANA top-level domains.
-     *  List accurate as of 2007/06/15.  List taken from:
+     *  List accurate as of 2010/02/05.  List taken from:
      *  http://data.iana.org/TLD/tlds-alpha-by-domain.txt
-     *  This pattern is auto-generated by //device/tools/make-iana-tld-pattern.py
+     *  This pattern is auto-generated by development/tools/make-iana-tld-pattern.py
      */
-    public static final Pattern TOP_LEVEL_DOMAIN
-        = Pattern.compile(
-                "((aero|arpa|asia|a[cdefgilmnoqrstuwxz])"
-                + "|(biz|b[abdefghijmnorstvwyz])"
-                + "|(cat|com|coop|c[acdfghiklmnoruvxyz])"
-                + "|d[ejkmoz]"
-                + "|(edu|e[cegrstu])"
-                + "|f[ijkmor]"
-                + "|(gov|g[abdefghilmnpqrstuwy])"
-                + "|h[kmnrtu]"
-                + "|(info|int|i[delmnoqrst])"
-                + "|(jobs|j[emop])"
-                + "|k[eghimnrwyz]"
-                + "|l[abcikrstuvy]"
-                + "|(mil|mobi|museum|m[acdghklmnopqrstuvwxyz])"
-                + "|(name|net|n[acefgilopruz])"
-                + "|(org|om)"
-                + "|(pro|p[aefghklmnrstwy])"
-                + "|qa"
-                + "|r[eouw]"
-                + "|s[abcdeghijklmnortuvyz]"
-                + "|(tel|travel|t[cdfghjklmnoprtvwz])"
-                + "|u[agkmsyz]"
-                + "|v[aceginu]"
-                + "|w[fs]"
-                + "|y[etu]"
-                + "|z[amw])");
+    public static final Pattern TOP_LEVEL_DOMAIN = Pattern.compile(
+        "((aero|arpa|asia|a[cdefgilmnoqrstuwxz])"
+        + "|(biz|b[abdefghijmnorstvwyz])"
+        + "|(cat|com|coop|c[acdfghiklmnoruvxyz])"
+        + "|d[ejkmoz]"
+        + "|(edu|e[cegrstu])"
+        + "|f[ijkmor]"
+        + "|(gov|g[abdefghilmnpqrstuwy])"
+        + "|h[kmnrtu]"
+        + "|(info|int|i[delmnoqrst])"
+        + "|(jobs|j[emop])"
+        + "|k[eghimnprwyz]"
+        + "|l[abcikrstuvy]"
+        + "|(mil|mobi|museum|m[acdeghklmnopqrstuvwxyz])"
+        + "|(name|net|n[acefgilopruz])"
+        + "|(org|om)"
+        + "|(pro|p[aefghklmnrstwy])"
+        + "|qa"
+        + "|r[eosuw]"
+        + "|s[abcdeghijklmnortuvyz]"
+        + "|(tel|travel|t[cdfghjklmnoprtvwz])"
+        + "|u[agksyz]"
+        + "|v[aceginu]"
+        + "|w[fs]"
+        + "|(xn\\-\\-0zwm56d|xn\\-\\-11b5bs3a9aj6g|xn\\-\\-80akhbyknj4f|xn\\-\\-9t4b11yi5a|xn\\-\\-deba0ad|xn\\-\\-g6w251d|xn\\-\\-hgbk6aj7f53bba|xn\\-\\-hlcj6aya9esc7a|xn\\-\\-jxalpdlp|xn\\-\\-kgbechtv|xn\\-\\-zckzah)"
+        + "|y[etu]"
+        + "|z[amw])");
 
     /**
      *  Regular expression pattern to match RFC 1738 URLs
-     *  List accurate as of 2007/06/15.  List taken from:
+     *  List accurate as of 2010/02/05.  List taken from:
      *  http://data.iana.org/TLD/tlds-alpha-by-domain.txt
-     *  This pattern is auto-generated by //device/tools/make-iana-tld-pattern.py
+     *  This pattern is auto-generated by development/tools/make-iana-tld-pattern.py
      */
-    public static final Pattern WEB_URL
-        = Pattern.compile(
-            "((?:(http|https|Http|Https|rtsp|Rtsp):\\/\\/(?:(?:[a-zA-Z0-9\\$\\-\\_\\.\\+\\!\\*\\'\\(\\)"
-            + "\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,64}(?:\\:(?:[a-zA-Z0-9\\$\\-\\_"
-            + "\\.\\+\\!\\*\\'\\(\\)\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,25})?\\@)?)?"
-            + "((?:(?:[a-zA-Z0-9][a-zA-Z0-9\\-]{0,64}\\.)+"   // named host
-            + "(?:"   // plus top level domain
-            + "(?:aero|arpa|asia|a[cdefgilmnoqrstuwxz])"
-            + "|(?:biz|b[abdefghijmnorstvwyz])"
-            + "|(?:cat|com|coop|c[acdfghiklmnoruvxyz])"
-            + "|d[ejkmoz]"
-            + "|(?:edu|e[cegrstu])"
-            + "|f[ijkmor]"
-            + "|(?:gov|g[abdefghilmnpqrstuwy])"
-            + "|h[kmnrtu]"
-            + "|(?:info|int|i[delmnoqrst])"
-            + "|(?:jobs|j[emop])"
-            + "|k[eghimnrwyz]"
-            + "|l[abcikrstuvy]"
-            + "|(?:mil|mobi|museum|m[acdghklmnopqrstuvwxyz])"
-            + "|(?:name|net|n[acefgilopruz])"
-            + "|(?:org|om)"
-            + "|(?:pro|p[aefghklmnrstwy])"
-            + "|qa"
-            + "|r[eouw]"
-            + "|s[abcdeghijklmnortuvyz]"
-            + "|(?:tel|travel|t[cdfghjklmnoprtvwz])"
-            + "|u[agkmsyz]"
-            + "|v[aceginu]"
-            + "|w[fs]"
-            + "|y[etu]"
-            + "|z[amw]))"
-            + "|(?:(?:25[0-5]|2[0-4]" // or ip address
-            + "[0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9])\\.(?:25[0-5]|2[0-4][0-9]"
-            + "|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(?:25[0-5]|2[0-4][0-9]|[0-1]"
-            + "[0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(?:25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}"
-            + "|[1-9][0-9]|[0-9])))"
-            + "(?:\\:\\d{1,5})?)" // plus option port number
-            + "(\\/(?:(?:[a-zA-Z0-9\\;\\/\\?\\:\\@\\&\\=\\#\\~"  // plus option query params
-            + "\\-\\.\\+\\!\\*\\'\\(\\)\\,\\_])|(?:\\%[a-fA-F0-9]{2}))*)?"
-            + "(?:\\b|$)"); // and finally, a word boundary or end of
-                            // input.  This is to stop foo.sure from
-                            // matching as foo.su
+    public static final Pattern WEB_URL = Pattern.compile(
+        "((?:(http|https|Http|Https|rtsp|Rtsp):\\/\\/(?:(?:[a-zA-Z0-9\\$\\-\\_\\.\\+\\!\\*\\'\\(\\)"
+        + "\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,64}(?:\\:(?:[a-zA-Z0-9\\$\\-\\_"
+        + "\\.\\+\\!\\*\\'\\(\\)\\,\\;\\?\\&\\=]|(?:\\%[a-fA-F0-9]{2})){1,25})?\\@)?)?"
+        + "((?:(?:[a-zA-Z0-9][a-zA-Z0-9\\-]{0,64}\\.)+"   // named host
+        + "(?:"   // plus top level domain
+        + "(?:aero|arpa|asia|a[cdefgilmnoqrstuwxz])"
+        + "|(?:biz|b[abdefghijmnorstvwyz])"
+        + "|(?:cat|com|coop|c[acdfghiklmnoruvxyz])"
+        + "|d[ejkmoz]"
+        + "|(?:edu|e[cegrstu])"
+        + "|f[ijkmor]"
+        + "|(?:gov|g[abdefghilmnpqrstuwy])"
+        + "|h[kmnrtu]"
+        + "|(?:info|int|i[delmnoqrst])"
+        + "|(?:jobs|j[emop])"
+        + "|k[eghimnprwyz]"
+        + "|l[abcikrstuvy]"
+        + "|(?:mil|mobi|museum|m[acdeghklmnopqrstuvwxyz])"
+        + "|(?:name|net|n[acefgilopruz])"
+        + "|(?:org|om)"
+        + "|(?:pro|p[aefghklmnrstwy])"
+        + "|qa"
+        + "|r[eosuw]"
+        + "|s[abcdeghijklmnortuvyz]"
+        + "|(?:tel|travel|t[cdfghjklmnoprtvwz])"
+        + "|u[agksyz]"
+        + "|v[aceginu]"
+        + "|w[fs]"
+        + "|(?:xn\\-\\-0zwm56d|xn\\-\\-11b5bs3a9aj6g|xn\\-\\-80akhbyknj4f|xn\\-\\-9t4b11yi5a|xn\\-\\-deba0ad|xn\\-\\-g6w251d|xn\\-\\-hgbk6aj7f53bba|xn\\-\\-hlcj6aya9esc7a|xn\\-\\-jxalpdlp|xn\\-\\-kgbechtv|xn\\-\\-zckzah)"
+        + "|y[etu]"
+        + "|z[amw]))"
+        + "|(?:(?:25[0-5]|2[0-4]" // or ip address
+        + "[0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9])\\.(?:25[0-5]|2[0-4][0-9]"
+        + "|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(?:25[0-5]|2[0-4][0-9]|[0-1]"
+        + "[0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(?:25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}"
+        + "|[1-9][0-9]|[0-9])))"
+        + "(?:\\:\\d{1,5})?)" // plus option port number
+        + "(\\/(?:(?:[a-zA-Z0-9\\;\\/\\?\\:\\@\\&\\=\\#\\~"  // plus option query params
+        + "\\-\\.\\+\\!\\*\\'\\(\\)\\,\\_])|(?:\\%[a-fA-F0-9]{2}))*)?"
+        + "(?:\\b|$)"); // and finally, a word boundary or end of
+                        // input.  This is to stop foo.sure from
+                        // matching as foo.su
 
     public static final Pattern IP_ADDRESS
         = Pattern.compile(
diff --git a/common/tests/src/com/android/common/PatternsTest.java b/common/tests/src/com/android/common/PatternsTest.java
index 7fabe5e..635601e 100644
--- a/common/tests/src/com/android/common/PatternsTest.java
+++ b/common/tests/src/com/android/common/PatternsTest.java
@@ -31,6 +31,20 @@ public class PatternsTest extends TestCase {
         t = Patterns.TOP_LEVEL_DOMAIN.matcher("com").matches();
         assertTrue("Missed valid TLD", t);
 
+        // One of the new top level domain.
+        t = Patterns.TOP_LEVEL_DOMAIN.matcher("me").matches();
+        assertTrue("Missed valid TLD", t);
+
+        // One of the new top level test domain.
+        t = Patterns.TOP_LEVEL_DOMAIN.matcher("xn--0zwm56d").matches();
+        assertTrue("Missed valid TLD", t);
+
+        t = Patterns.TOP_LEVEL_DOMAIN.matcher("mem").matches();
+        assertFalse("Matched invalid TLD!", t);
+
+        t = Patterns.TOP_LEVEL_DOMAIN.matcher("xn").matches();
+        assertFalse("Matched invalid TLD!", t);
+
         t = Patterns.TOP_LEVEL_DOMAIN.matcher("xer").matches();
         assertFalse("Matched invalid TLD!", t);
     }
@@ -42,6 +56,18 @@ public class PatternsTest extends TestCase {
         t = Patterns.WEB_URL.matcher("http://www.google.com").matches();
         assertTrue("Valid URL", t);
 
+        // Google in one of the new top level domain.
+        t = Patterns.WEB_URL.matcher("http://www.google.me").matches();
+        assertTrue("Valid URL", t);
+        t = Patterns.WEB_URL.matcher("google.me").matches();
+        assertTrue("Valid URL", t);
+
+        // Test url in Chinese: http://xn--fsqu00a.xn--0zwm56d
+        t = Patterns.WEB_URL.matcher("http://xn--fsqu00a.xn--0zwm56d").matches();
+        assertTrue("Valid URL", t);
+        t = Patterns.WEB_URL.matcher("xn--fsqu00a.xn--0zwm56d").matches();
+        assertTrue("Valid URL", t);
+
         t = Patterns.WEB_URL.matcher("ftp://www.example.com").matches();
         assertFalse("Matched invalid protocol", t);