jars/icu4j-4_4_2-src/main/tests/core/src/com/ibm/icu/dev/test/normalizer/NormalizationMonkeyTest.java

   1 /*\r
   2  *******************************************************************************\r
   3  * Copyright (C) 1996-2004, International Business Machines Corporation and    *\r
   4  * others. All Rights Reserved.                                                *\r
   5  *******************************************************************************\r
   6  */\r
   7 \r
   8 package com.ibm.icu.dev.test.normalizer;\r
   9 \r
  10 import java.util.Random;\r
  11 \r
  12 import com.ibm.icu.dev.test.TestFmwk;\r
  13 import com.ibm.icu.lang.UCharacter;\r
  14 import com.ibm.icu.lang.UCharacterCategory;\r
  15 import com.ibm.icu.text.Normalizer;\r
  16 import com.ibm.icu.text.UTF16;\r
  17 \r
  18 public class NormalizationMonkeyTest extends TestFmwk {\r
  19     int loopCount = 100;\r
  20     int maxCharCount = 20;\r
  21     int maxCodePoint = 0x10ffff;\r
  22     Random random = null; // initialized in getTestSource\r
  23     UnicodeNormalizer unicode_NFD;\r
  24     UnicodeNormalizer unicode_NFC;\r
  25     UnicodeNormalizer unicode_NFKD;\r
  26     UnicodeNormalizer unicode_NFKC;\r
  27     \r
  28     public NormalizationMonkeyTest() {\r
  29     }\r
  30     \r
  31     public static void main(String[] args) throws Exception {\r
  32         new NormalizationMonkeyTest().run(args);\r
  33     }\r
  34     \r
  35     public void TestNormalize() {\r
  36         if (unicode_NFD == null) {\r
  37             try {\r
  38                 unicode_NFD = new UnicodeNormalizer(UnicodeNormalizer.D, true);\r
  39                 unicode_NFC = new UnicodeNormalizer(UnicodeNormalizer.C, true);\r
  40                 unicode_NFKD = new UnicodeNormalizer(UnicodeNormalizer.KD, true);\r
  41                 unicode_NFKC = new UnicodeNormalizer(UnicodeNormalizer.KC, true);\r
  42             }\r
  43             catch (Exception e) {\r
  44                 String msg = "Normalization tests could not be run: " + e.getMessage();\r
  45                 if (isModularBuild()) {\r
  46                     warnln(msg);\r
  47                 } else {\r
  48                     errln(msg);\r
  49                 }\r
  50             }\r
  51         }\r
  52         int i = 0;\r
  53         while (i < loopCount) {\r
  54             String source = getTestSource();\r
  55             logln("Test source:" + source);\r
  56             //NFD\r
  57             String uncodeNorm = unicode_NFD.normalize(source);\r
  58             String icuNorm = Normalizer.normalize(source, Normalizer.NFD);\r
  59             logln("\tNFD(Unicode): " + uncodeNorm);\r
  60             logln("\tNFD(icu4j)  : " + icuNorm);\r
  61             if (!uncodeNorm.equals(icuNorm)) {\r
  62                 errln("NFD: Unicode sample output => " + uncodeNorm + "; icu4j output=> " + icuNorm);\r
  63             }\r
  64             //NFC\r
  65             uncodeNorm = unicode_NFC.normalize(source);\r
  66             icuNorm = Normalizer.normalize(source, Normalizer.NFC);\r
  67             logln("\tNFC(Unicode): " + uncodeNorm);\r
  68             logln("\tNFC(icu4j)  : " + icuNorm);\r
  69             if (!uncodeNorm.equals(icuNorm)) {\r
  70                 errln("NFC: Unicode sample output => " + uncodeNorm + "; icu4j output=> " + icuNorm);\r
  71             }\r
  72             //NFKD\r
  73             uncodeNorm = unicode_NFKD.normalize(source);\r
  74             icuNorm = Normalizer.normalize(source, Normalizer.NFKD);\r
  75             logln("\tNFKD(Unicode): " + uncodeNorm);\r
  76             logln("\tNFKD(icu4j)  : " + icuNorm);\r
  77             if (!uncodeNorm.equals(icuNorm)) {\r
  78                 errln("NFKD: Unicode sample output => " + uncodeNorm + "; icu4j output=> " + icuNorm);\r
  79             }\r
  80             //NFKC\r
  81             uncodeNorm = unicode_NFKC.normalize(source);\r
  82             icuNorm = Normalizer.normalize(source, Normalizer.NFKC);\r
  83             logln("\tNFKC(Unicode): " + uncodeNorm);\r
  84             logln("\tNFKC(icu4j)  : " + icuNorm);\r
  85             if (!uncodeNorm.equals(icuNorm)) {\r
  86                 errln("NFKC: Unicode sample output => " + uncodeNorm + "; icu4j output=> " + icuNorm);\r
  87             }\r
  88             \r
  89             i++;\r
  90         }\r
  91     }\r
  92     \r
  93     String getTestSource() {\r
  94     if (random == null) {\r
  95         random = createRandom(); // use test framework's random seed\r
  96     }\r
  97         String source = "";\r
  98         int i = 0;\r
  99         while (i < (random.nextInt(maxCharCount) + 1)) {\r
 100             int codepoint = random.nextInt(maxCodePoint);\r
 101             //Elimate unassigned characters\r
 102             while (UCharacter.getType(codepoint) == UCharacterCategory.UNASSIGNED) {\r
 103                 codepoint = random.nextInt(maxCodePoint);\r
 104             }\r
 105             source = source + UTF16.valueOf(codepoint);\r
 106             i++;\r
 107         }\r
 108         return source;\r
 109     }\r
 110 }\r