Minor cleanup

Author: macchiati

unicodetools/data/linkification/dev/LinkBracket.txt

@@ -1,5 +1,5 @@
 # LinkBracket.txt
-# Date: 2025-12-11, 02:19:40 GMT 
+# Date: 2025-12-12, 00:48:36 GMT 
 # © 2025 Unicode®, Inc.
 # Unicode and the Unicode Logo are registered trademarks of Unicode, Inc. in the U.S. and other countries.
 # For terms of use and license, see https://www.unicode.org/terms_of_use.html

unicodetools/data/linkification/dev/LinkDetectionTest.txt

@@ -1,5 +1,5 @@
 # LinkDetectionTest.txt
-# Date: 2025-12-11, 02:19:40 GMT 
+# Date: 2025-12-12, 00:48:36 GMT 
 # © 2025 Unicode®, Inc.
 # Unicode and the Unicode Logo are registered trademarks of Unicode, Inc. in the U.S. and other countries.
 # For terms of use and license, see https://www.unicode.org/terms_of_use.html

unicodetools/data/linkification/dev/LinkEmail.txt

@@ -1,5 +1,5 @@
 # LinkEmail.txt
-# Date: 2025-12-11, 02:19:40 GMT 
+# Date: 2025-12-12, 00:48:36 GMT 
 # © 2025 Unicode®, Inc.
 # Unicode and the Unicode Logo are registered trademarks of Unicode, Inc. in the U.S. and other countries.
 # For terms of use and license, see https://www.unicode.org/terms_of_use.html

unicodetools/data/linkification/dev/LinkFormattingTest.txt

@@ -1,5 +1,5 @@
 # LinkFormattingTest.txt
-# Date: 2025-12-11, 02:19:40 GMT 
+# Date: 2025-12-12, 00:48:36 GMT 
 # © 2025 Unicode®, Inc.
 # Unicode and the Unicode Logo are registered trademarks of Unicode, Inc. in the U.S. and other countries.
 # For terms of use and license, see https://www.unicode.org/terms_of_use.html

unicodetools/data/linkification/dev/LinkTerm.txt

@@ -1,5 +1,5 @@
 # LinkTerm.txt
-# Date: 2025-12-11, 02:19:40 GMT 
+# Date: 2025-12-12, 00:48:36 GMT 
 # © 2025 Unicode®, Inc.
 # Unicode and the Unicode Logo are registered trademarks of Unicode, Inc. in the U.S. and other countries.
 # For terms of use and license, see https://www.unicode.org/terms_of_use.html

unicodetools/src/main/java/org/unicode/tools/GenerateLinkData.java

@@ -196,6 +196,7 @@ static void writeTestHeader(
 
     /** Generate property data for the UTS */
     static void generatePropertyData() {
+        System.out.println("TLDs=\t" + Joiner.on(' ').join(LinkUtilities.TLDS));
 
         BagFormatter bf = new BagFormatter(LinkUtilities.IUP).setLineSeparator("\n");
 
@@ -211,6 +212,7 @@ static void generatePropertyData() {
                 bf.showSetNames(out, propValue.base);
                 out.println("");
                 out.flush();
+                System.out.println(propValue + "=\t" + propValue.base.toPattern(false));
             }
 
         } catch (IOException e) {
@@ -237,7 +239,9 @@ static void generatePropertyData() {
                     "LinkEmail",
                     "Link_Email",
                     UcdPropertyValues.Binary.No.toString());
-            bf.showSetNames(out, LinkUtilities.LinkEmail.getSet(UcdPropertyValues.Binary.Yes));
+            UnicodeSet linkEmailSet = LinkUtilities.LinkEmail.getSet(UcdPropertyValues.Binary.Yes);
+            bf.showSetNames(out, linkEmailSet);
+            System.out.println("LinkEmail=\t" + linkEmailSet.toPattern(false));
         } catch (IOException e) {
             throw new UncheckedIOException(e);
         }

unicodetools/src/main/java/org/unicode/utilities/LinkUtilities.java

@@ -5,13 +5,15 @@
 import com.google.common.collect.ComparisonChain;
 import com.google.common.collect.ImmutableSet;
 import com.google.common.collect.ImmutableSortedMap;
+import com.google.common.collect.ImmutableSortedSet;
 import com.google.common.collect.Sets;
 import com.google.common.util.concurrent.UncheckedExecutionException;
 import com.ibm.icu.impl.IDNA2003;
 import com.ibm.icu.impl.UnicodeMap;
 import com.ibm.icu.lang.UCharacter;
 import com.ibm.icu.lang.UProperty;
 import com.ibm.icu.lang.UProperty.NameChoice;
+import com.ibm.icu.text.Collator;
 import com.ibm.icu.text.StringPrepParseException;
 import com.ibm.icu.text.UTF16;
 import com.ibm.icu.text.UnicodeSet;
@@ -27,10 +29,12 @@
 import java.util.EnumSet;
 import java.util.HashMap;
 import java.util.List;
+import java.util.Locale;
 import java.util.Map;
 import java.util.Map.Entry;
 import java.util.NavigableMap;
 import java.util.Set;
+import java.util.SortedSet;
 import java.util.Stack;
 import java.util.TreeSet;
 import java.util.regex.Matcher;
@@ -50,8 +54,6 @@
 import org.unicode.utilities.LinkUtilities.LinkScanner;
 
 public class LinkUtilities {
-    private static final boolean SHOW_NON_ASCII_TLDS = true;
-
     // allow changing UnicodeSet to use the current IndexUnicodeProperties
     public static final IndexUnicodeProperties IUP =
             IndexUnicodeProperties.make(VersionInfo.UNICODE_17_0);
@@ -830,6 +832,8 @@ public void showLinkPairedOpeners() {
      */
     public static final Pattern TLD_SCANNER;
 
+    public static final SortedSet<String> TLDS;
+
     public static final String DOTSET_STRING = "[.。]";
     public static final UnicodeSet DOTSET = new UnicodeSet("[.。]").freeze();
     public static final Splitter SPLIT_LABELS = Splitter.on(Pattern.compile("[.。]"));
@@ -870,10 +874,13 @@ public int compare(String o1, String o2) {
                                 }
                             });
             String pattern = "(?u)" + DOTSET_STRING + "(" + Joiner.on('|').join(core) + ")";
+            TLDS =
+                    core.stream()
+                            .map(x -> UCharacter.toLowerCase(x))
+                            .collect(
+                                    ImmutableSortedSet.toImmutableSortedSet(
+                                            Collator.getInstance(Locale.ROOT)));
             TLD_SCANNER = Pattern.compile(pattern);
-            if (SHOW_NON_ASCII_TLDS) {
-                System.out.println(nonAscii);
-            }
         } catch (IOException e) {
             throw new UncheckedIOException(e);
         }