Fixing Break Charts & Tests

X-SVN-Rev: 11428

tools/unicodetools/com/ibm/text/UCA/WriteCollationData.java

@@ -5,8 +5,8 @@
 *******************************************************************************
 *
 * $Source: /xsrl/Nsvn/icu/unicodetools/com/ibm/text/UCA/WriteCollationData.java,v $ 
-* $Date: 2003/03/19 23:31:12 $ 
-* $Revision: 1.30 $
+* $Date: 2003/04/01 02:51:57 $ 
+* $Revision: 1.31 $
 *
 *******************************************************************************
 */
@@ -2492,17 +2492,26 @@ F900..FAFF; CJK Compatibility Ideographs
     
     static StringBuffer quoteOperandBuffer = new StringBuffer(); // faster
     
+    static UnicodeSet needsQuoting = null;
+    
     static final String quoteOperand(String s) {
+        if (needsQuoting == null) {
+            /*
+            c >= 'a' && c <= 'z' 
+              || c >= 'A' && c <= 'Z' 
+              || c >= '0' && c <= '9'
+              || (c >= 0xA0 && !UCharacterProperty.isRuleWhiteSpace(c))
+              */
+            needsQuoting = new UnicodeSet("[a-zA-Z0-9\\u00A0-\\U00010FFF]");
+            needsQuoting.remove();
+        }
     	s = NFC.normalize(s);
         quoteOperandBuffer.setLength(0);
         boolean noQuotes = true;
         boolean inQuote = false;
         for (int i = 0; i < s.length(); ++i) {
             char c = s.charAt(i);
-            if (c >= 'a' && c <= 'z' 
-              || c >= 'A' && c <= 'Z' 
-              || c >= '0' && c <= '9'
-              || (c >= 0xA0 && !UCharacterProperty.isRuleWhiteSpace(c))) {
+            if (!needsQuoting.contains(c)) {
                 if (inQuote) {
                     quoteOperandBuffer.append('\'');
                     inQuote = false;

tools/unicodetools/com/ibm/text/UCD/DerivedProperty.java

@@ -5,8 +5,8 @@
 *******************************************************************************
 *
 * $Source: /xsrl/Nsvn/icu/unicodetools/com/ibm/text/UCD/DerivedProperty.java,v $
-* $Date: 2003/03/12 16:01:26 $
-* $Revision: 1.19 $
+* $Date: 2003/04/01 02:52:00 $
+* $Revision: 1.20 $
 *
 *******************************************************************************
 */
@@ -501,7 +501,7 @@ public final class DerivedProperty implements UCD_Types {
             public boolean hasValue(int cp) {
                 byte cat = ucdData.getCategory(cp);
                 if (cat == Lu || cat == Ll || cat == Lt || cat == Lm || cat == Lo || cat == Nl
-                || ucdData.getBinaryProperty(cp, Alphabetic)) return true;
+                || ucdData.getBinaryProperty(cp, Other_Alphabetic)) return true;
                 return false;
             }
         };

tools/unicodetools/com/ibm/text/UCD/UCD_Types.java

@@ -5,8 +5,8 @@
 *******************************************************************************
 *
 * $Source: /xsrl/Nsvn/icu/unicodetools/com/ibm/text/UCD/UCD_Types.java,v $
-* $Date: 2003/03/20 21:47:26 $
-* $Revision: 1.21 $
+* $Date: 2003/04/01 02:52:00 $
+* $Revision: 1.22 $
 *
 *******************************************************************************
 */
@@ -190,7 +190,7 @@ public interface UCD_Types {
         Math_Property = 10,
         Hex_Digit = 11,
         ASCII_Hex_Digit = 12,
-	    Alphabetic = 13,
+	    Other_Alphabetic = 13,
         Ideographic = 14,
         Diacritic = 15,
         Extender = 16,
@@ -215,7 +215,7 @@ public interface UCD_Types {
     static final int
 	    BidiMirroredMask = 1<<BidiMirrored,
 	    CompositionExclusionMask = 1<<CompositionExclusion,
-	    AlphabeticMask = 1<<Alphabetic,
+	    AlphabeticMask = 1<<Other_Alphabetic,
 	    Bidi_ControlMask = 1<<Bidi_Control,
         DashMask = 1<<Dash,
         DiacriticMask = 1<<Diacritic,
@@ -466,7 +466,7 @@ public static byte
     
     // DERIVED PROPERTY
     
-    static final int
+    static final byte
         PropMath = 0,
         PropAlphabetic = 1,
         PropLowercase = 2,

tools/unicodetools/com/ibm/text/UCD/UnicodeMap.java

@@ -0,0 +1,94 @@
+/**
+*******************************************************************************
+* Copyright (C) 1996-2001, International Business Machines Corporation and    *
+* others. All Rights Reserved.                                                *
+*******************************************************************************
+*
+* $Source: /xsrl/Nsvn/icu/unicodetools/com/ibm/text/UCD/Attic/UnicodeMap.java,v $
+* $Date: 2003/04/01 02:53:07 $
+* $Revision: 1.1 $
+*
+*******************************************************************************
+*/
+
+package com.ibm.text.UCD;
+
+import java.util.*;
+import java.io.*;
+
+import com.ibm.text.utility.*;
+import com.ibm.icu.text.UTF16;
+import com.ibm.icu.text.UnicodeSet;
+
+/**
+ * Class that maps from codepoints to an index, and optionally a label.
+ */
+public class UnicodeMap {
+    UnicodeSet[] sets = new UnicodeSet[50];
+    String[] labels = new String[50];
+    int count = 0;
+    
+    public int add(String label, UnicodeSet set) {
+        return add(label, set, false, true);
+    }
+    
+    /**
+     * Add set
+     *@param removeOld true: remove any collisions from sets already in the map
+     * if false, remove any collisions from this set
+     *@param signal: print a warning when collisions occur
+     */
+    public int add(String label, UnicodeSet set, boolean removeOld, boolean signal) {
+        // remove from any preceding!!
+        for (int i = 0; i < count; ++i) {
+            if (!set.containsSome(sets[i])) continue;
+            if (signal) showOverlap(label, set, i);
+            if (removeOld) {
+                sets[i] = sets[i].removeAll(set);
+            } else {
+                set = set.removeAll(sets[i]);
+            }
+        }
+        sets[count] = set;
+        labels[count++] = label;
+        return (short)(count - 1);
+    }
+    
+    public void showOverlap(String label, UnicodeSet set, int i) {
+        UnicodeSet delta = new UnicodeSet(set).retainAll(sets[i]);
+        System.out.println("Warning! Overlap with " + label + " and " + labels[i]
+            + ": " + delta);
+    }
+    
+    public int getIndex(int codepoint) {
+        for (int i = count - 1; i >= 0; --i) {
+            if (sets[i].contains(codepoint)) return i;
+        }
+        return -1;
+    }
+    
+    public int getIndexFromLabel(String label) {
+        for (int i = count - 1; i >= 0; --i) {
+            if (labels[i].equalsIgnoreCase(label)) return i;
+        }
+        return -1;
+    }
+
+    public String getLabel(int codepoint) {
+        return getLabelFromIndex(getIndex(codepoint));
+    }
+
+    public String getLabelFromIndex(int index) {
+        if (index < 0 || index >= count) return null;
+        return labels[index];
+    }
+
+    public UnicodeSet getSetFromIndex(int index) {
+        if (index < 0 || index >= count) return null;
+        return new UnicodeSet(sets[index]); // protect from changes
+    }
+    
+    public int size() {
+        return count;
+    }
+}

tools/unicodetools/com/ibm/text/utility/Utility.java

@@ -5,8 +5,8 @@
 *******************************************************************************
 *
 * $Source: /xsrl/Nsvn/icu/unicodetools/com/ibm/text/utility/Utility.java,v $
-* $Date: 2003/03/19 17:30:56 $
-* $Revision: 1.30 $
+* $Date: 2003/04/01 02:52:00 $
+* $Revision: 1.31 $
 *
 *******************************************************************************
 */
@@ -1071,6 +1071,15 @@ public final class Utility implements UCD_Types {    // COMMON UTILITIES
     
     static PrintWriter showSetNamesPw;
     
+    public static void showSetDifferences(String name1, UnicodeSet set1, String name2, UnicodeSet set2, boolean separateLines, UCD ucd) {
+        UnicodeSet temp = new UnicodeSet(set1).removeAll(set2);
+        showSetNames("In " + name1 + ", but not " + name2,  temp,  separateLines,  false,  false, ucd);
+        temp = new UnicodeSet(set2).removeAll(set1);
+        showSetNames("In " + name2 + ", but not " + name1,  temp,  separateLines,  false,  false, ucd);
+        temp = new UnicodeSet(set2).retainAll(set1);
+        showSetNames("In " + name1 + " and " + name2,  temp,  separateLines,  false,  false, ucd);
+    }
+    
     public static void showSetNames(String prefix, UnicodeSet set, boolean separateLines, UCD ucd) {
         showSetNames(prefix,  set,  separateLines,  false,  false, ucd);
     }