michael@0: /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
michael@0: /* This Source Code Form is subject to the terms of the Mozilla Public
michael@0:  * License, v. 2.0. If a copy of the MPL was not distributed with this
michael@0:  * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
michael@0: 
michael@0: #include <stdio.h>
michael@0: #include "nsXPCOM.h"
michael@0: #include "nsIEntityConverter.h"
michael@0: #include "nsISaveAsCharset.h"
michael@0: #include "nsCOMPtr.h"
michael@0: #include "nsIUnicodeNormalizer.h"
michael@0: #include "nsStringAPI.h"
michael@0: #include "nsUnicharUtils.h"
michael@0: #include "nsMemory.h"
michael@0: #include "nsComponentManagerUtils.h"
michael@0: #include "nsServiceManagerUtils.h"
michael@0: 
michael@0: NS_DEFINE_CID(kEntityConverterCID, NS_ENTITYCONVERTER_CID);
michael@0: NS_DEFINE_CID(kSaveAsCharsetCID, NS_SAVEASCHARSET_CID);
michael@0: NS_DEFINE_CID(kUnicodeNormalizerCID, NS_UNICODE_NORMALIZER_CID);
michael@0: 
michael@0: #define TESTLEN 32
michael@0: #define T2LEN TESTLEN
michael@0: #define T3LEN TESTLEN
michael@0: #define T4LEN TESTLEN
michael@0: 
michael@0: // test data for ToUpper 
michael@0: static char16_t t2data  [T2LEN+1] = {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0043 ,  //  2
michael@0:   0x0067 ,  //  3
michael@0:   0x00C8 ,  //  4
michael@0:   0x00E9 ,  //  5
michael@0:   0x0147 ,  //  6
michael@0:   0x01C4 ,  //  7
michael@0:   0x01C6 ,  //  8
michael@0:   0x01C5 ,  //  9
michael@0:   0x03C0 ,  // 10
michael@0:   0x03B2 ,  // 11
michael@0:   0x0438 ,  // 12
michael@0:   0x04A5 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5185 ,  // 17
michael@0:   0xC021 ,  // 18
michael@0:   0xFF48 ,  // 19
michael@0:   0x01C7 ,  // 20
michael@0:   0x01C8 ,  // 21
michael@0:   0x01C9 ,  // 22
michael@0:   0x01CA ,  // 23
michael@0:   0x01CB ,  // 24
michael@0:   0x01CC ,  // 25
michael@0:   0x01F1 ,  // 26
michael@0:   0x01F2 ,  // 27
michael@0:   0x01F3 ,  // 28
michael@0:   0x0250 ,  // 29
michael@0:   0x0271 ,  // 30
michael@0:   0xA641 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: // expected result for ToUpper 
michael@0: static char16_t t2result[T2LEN+1] =  {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0043 ,  //  2
michael@0:   0x0047 ,  //  3
michael@0:   0x00C8 ,  //  4
michael@0:   0x00C9 ,  //  5
michael@0:   0x0147 ,  //  6
michael@0:   0x01C4 ,  //  7
michael@0:   0x01C4 ,  //  8
michael@0:   0x01C4 ,  //  9
michael@0:   0x03A0 ,  // 10
michael@0:   0x0392 ,  // 11
michael@0:   0x0418 ,  // 12
michael@0:   0x04A4 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5185 ,  // 17
michael@0:   0xC021 ,  // 18
michael@0:   0xFF28 ,  // 19
michael@0:   0x01C7 ,  // 20
michael@0:   0x01C7 ,  // 21
michael@0:   0x01C7 ,  // 22
michael@0:   0x01CA ,  // 23
michael@0:   0x01CA ,  // 24
michael@0:   0x01CA ,  // 25
michael@0:   0x01F1 ,  // 26
michael@0:   0x01F1 ,  // 27
michael@0:   0x01F1 ,  // 28
michael@0:   0x2C6F ,  // 29
michael@0:   0x2C6E ,  // 30
michael@0:   0xA640 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: // test data for ToLower 
michael@0: static char16_t t3data  [T3LEN+1] =  {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0043 ,  //  2
michael@0:   0x0067 ,  //  3
michael@0:   0x00C8 ,  //  4
michael@0:   0x00E9 ,  //  5
michael@0:   0x0147 ,  //  6
michael@0:   0x01C4 ,  //  7
michael@0:   0x01C6 ,  //  8
michael@0:   0x01C5 ,  //  9
michael@0:   0x03A0 ,  // 10
michael@0:   0x0392 ,  // 11
michael@0:   0x0418 ,  // 12
michael@0:   0x04A4 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5187 ,  // 17
michael@0:   0xC023 ,  // 18
michael@0:   0xFF28 ,  // 19
michael@0:   0x01C7 ,  // 20
michael@0:   0x01C8 ,  // 21
michael@0:   0x01C9 ,  // 22
michael@0:   0x01CA ,  // 23
michael@0:   0x01CB ,  // 24
michael@0:   0x01CC ,  // 25
michael@0:   0x01F1 ,  // 26
michael@0:   0x01F2 ,  // 27
michael@0:   0x01F3 ,  // 28
michael@0:   0x2C6F ,  // 29
michael@0:   0x2C6E ,  // 30
michael@0:   0xA640 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: // expected result for ToLower 
michael@0: static char16_t t3result[T3LEN+1] =  {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0063 ,  //  2
michael@0:   0x0067 ,  //  3
michael@0:   0x00E8 ,  //  4
michael@0:   0x00E9 ,  //  5
michael@0:   0x0148 ,  //  6
michael@0:   0x01C6 ,  //  7
michael@0:   0x01C6 ,  //  8
michael@0:   0x01C6 ,  //  9
michael@0:   0x03C0 ,  // 10
michael@0:   0x03B2 ,  // 11
michael@0:   0x0438 ,  // 12
michael@0:   0x04A5 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5187 ,  // 17
michael@0:   0xC023 ,  // 18
michael@0:   0xFF48 ,  // 19
michael@0:   0x01C9 ,  // 20
michael@0:   0x01C9 ,  // 21
michael@0:   0x01C9 ,  // 22
michael@0:   0x01CC ,  // 23
michael@0:   0x01CC ,  // 24
michael@0:   0x01CC ,  // 25
michael@0:   0x01F3 ,  // 26
michael@0:   0x01F3 ,  // 27
michael@0:   0x01F3 ,  // 28
michael@0:   0x0250 ,  // 29
michael@0:   0x0271 ,  // 30
michael@0:   0xA641 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: // test data for ToTitle 
michael@0: static char16_t t4data  [T4LEN+2] =  {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0043 ,  //  2
michael@0:   0x0067 ,  //  3
michael@0:   0x00C8 ,  //  4
michael@0:   0x00E9 ,  //  5
michael@0:   0x0147 ,  //  6
michael@0:   0x01C4 ,  //  7
michael@0:   0x01C6 ,  //  8
michael@0:   0x01C5 ,  //  9
michael@0:   0x03C0 ,  // 10
michael@0:   0x03B2 ,  // 11
michael@0:   0x0438 ,  // 12
michael@0:   0x04A5 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5189 ,  // 17
michael@0:   0xC013 ,  // 18
michael@0:   0xFF52 ,  // 19
michael@0:   0x01C7 ,  // 20
michael@0:   0x01C8 ,  // 21
michael@0:   0x01C9 ,  // 22
michael@0:   0x01CA ,  // 23
michael@0:   0x01CB ,  // 24
michael@0:   0x01CC ,  // 25
michael@0:   0x01F1 ,  // 26
michael@0:   0x01F2 ,  // 27
michael@0:   0x01F3 ,  // 28
michael@0:   0x0250 ,  // 29
michael@0:   0x0271 ,  // 30
michael@0:   0xA641 ,  // 31
michael@0:   0x0041 ,  // Dummy entry to prevent overflow
michael@0:   0x00  
michael@0: };
michael@0: // expected result for ToTitle 
michael@0: static char16_t t4result[T4LEN+2] =  {
michael@0:   0x0031 ,  //  0
michael@0:   0x0019 ,  //  1
michael@0:   0x0043 ,  //  2
michael@0:   0x0047 ,  //  3
michael@0:   0x00C8 ,  //  4
michael@0:   0x00C9 ,  //  5
michael@0:   0x0147 ,  //  6
michael@0:   0x01C4 ,  //  7
michael@0:   0x01C5 ,  //  8
michael@0:   0x01C5 ,  //  9
michael@0:   0x03A0 ,  // 10
michael@0:   0x0392 ,  // 11
michael@0:   0x0418 ,  // 12
michael@0:   0x04A4 ,  // 13
michael@0:   0x05D0 ,  // 14
michael@0:   0x0A20 ,  // 15
michael@0:   0x30B0 ,  // 16
michael@0:   0x5189 ,  // 17
michael@0:   0xC013 ,  // 18
michael@0:   0xFF32 ,  // 19
michael@0:   0x01C7 ,  // 20
michael@0:   0x01C8 ,  // 21
michael@0:   0x01C8 ,  // 22
michael@0:   0x01CA ,  // 23
michael@0:   0x01CB ,  // 24
michael@0:   0x01CB ,  // 25
michael@0:   0x01F1 ,  // 26
michael@0:   0x01F2 ,  // 27
michael@0:   0x01F2 ,  // 28
michael@0:   0x2C6F ,  // 29
michael@0:   0x2C6E ,  // 30
michael@0:   0xA640 ,  // 31
michael@0:   0x0041 ,  // Dummy entry to prevent overflow
michael@0:   0x00  
michael@0: };
michael@0:  
michael@0: static unsigned char t6lhs[] = {
michael@0:   0x31 ,       //  0
michael@0:   0x19 ,       //  1
michael@0:   0x43 ,       //  2
michael@0:   0x67 ,       //  3
michael@0:   0xC3, 0x88 , //  4
michael@0:   0xC3, 0xA9 , //  5
michael@0:   0xC5, 0x87 , //  6
michael@0:   0xC7, 0x84 , //  7
michael@0:   0xC7, 0x86 , //  8
michael@0:   0xC7, 0x85 , //  9
michael@0:   0xCF, 0x80 ,  // 10
michael@0:   0xCE, 0xB2 ,  // 11
michael@0:   0xD0, 0xB8 ,  // 12
michael@0:   0xD2, 0xA5 ,  // 13
michael@0:   0xD7, 0x90 ,  // 14
michael@0:   0xE0, 0xA8, 0xA0 ,  // 15
michael@0:   0xE3, 0x82, 0xB0 ,  // 16
michael@0:   0xE5, 0x86, 0x85 ,  // 17
michael@0:   0xEC, 0x80, 0xA1 ,  // 18
michael@0:   0xEF, 0xBD, 0x88 ,  // 19
michael@0:   0xC7, 0x87 ,  // 20
michael@0:   0xC7, 0x88 ,  // 21
michael@0:   0xC7, 0x89 ,  // 22
michael@0:   0xC7, 0x8A ,  // 23
michael@0:   0xC7, 0x8B ,  // 24
michael@0:   0xC7, 0x8C ,  // 25
michael@0:   0xC7, 0xB1 ,  // 26
michael@0:   0xC7, 0xB2 ,  // 27
michael@0:   0xC7, 0xB3 ,  // 28
michael@0:   0xC9, 0x90 ,  // 29
michael@0:   0xC9, 0xB1 ,  // 30
michael@0:   0xEA, 0x99, 0x81 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: 
michael@0: static unsigned char t6rhs[] =  {
michael@0:   0x31 ,  //  0
michael@0:   0x19 ,  //  1
michael@0:   0x43 ,  //  2
michael@0:   0x47 ,  //  3
michael@0:   0xC3, 0x88 ,  //  4
michael@0:   0xC3, 0x89 ,  //  5
michael@0:   0xC5, 0x87 ,  //  6
michael@0:   0xC7, 0x84 ,  //  7
michael@0:   0xC7, 0x84 ,  //  8
michael@0:   0xC7, 0x84 ,  //  9
michael@0:   0xCE, 0xA0 ,  // 10
michael@0:   0xCE, 0x92 ,  // 11
michael@0:   0xD0, 0x98 ,  // 12
michael@0:   0xD2, 0xA4 ,  // 13
michael@0:   0xD7, 0x90 ,  // 14
michael@0:   0xE0, 0xA8, 0xA0 ,  // 15
michael@0:   0xE3, 0x82, 0xB0 ,  // 16
michael@0:   0xE5, 0x86, 0x85 ,  // 17
michael@0:   0xEC, 0x80, 0xA1 ,  // 18
michael@0:   0xEF, 0xBC, 0xA8 ,  // 19
michael@0:   0xC7, 0x87 ,  // 20
michael@0:   0xC7, 0x87 ,  // 21
michael@0:   0xC7, 0x87 ,  // 22
michael@0:   0xC7, 0x8a ,  // 23
michael@0:   0xC7, 0x8a ,  // 24
michael@0:   0xC7, 0x8a ,  // 25
michael@0:   0xC7, 0xB1 ,  // 26
michael@0:   0xC7, 0xB1 ,  // 27
michael@0:   0xC7, 0xB1 ,  // 28
michael@0:   0xE2, 0xB1, 0xAF ,  // 29
michael@0:   0xE2, 0xB1, 0xAE ,  // 30
michael@0:   0xEA, 0x99, 0x80 ,  // 31
michael@0:   0x00  
michael@0: };
michael@0: 
michael@0: static const char *t7lhs = "aBcDeFGHIJKL1!!2!!a!uuuu";
michael@0: static const char *t7rhs = "AbCdEFghijkL1!!2!!A!UUuU";
michael@0: 
michael@0: static const char *t8lhs = "aazzz";
michael@0: static const char *t8rhs = "aBa";
michael@0: 
michael@0: static const char *t9lhs = "@a";
michael@0: static const char *t9rhs = "`a";
michael@0: 
michael@0: bool CharByCharCompareEqual(const char *a, const char *b,
michael@0:                             uint32_t aLen, uint32_t bLen)
michael@0: {
michael@0:   // Do basically a CaseInsensitiveCompare(), but using
michael@0:   // CaseInsensitiveUTF8CharsEqual().
michael@0: 
michael@0:   const char *aEnd = a + aLen;
michael@0:   const char *bEnd = b + bLen;
michael@0:   while (a < aEnd && b < bEnd) {
michael@0:     bool err;
michael@0:     if (!CaseInsensitiveUTF8CharsEqual(a, b, aEnd, bEnd, &a, &b, &err) || err)
michael@0:       return false;
michael@0:   }
michael@0:   return true;
michael@0: }
michael@0: 
michael@0: void TestCaseConversion()
michael@0: {
michael@0:   printf("==========================\n");
michael@0:   printf("Start case conversion test\n");
michael@0:   printf("==========================\n");
michael@0: 
michael@0:   int i;
michael@0:   char16_t buf[256];
michael@0: 
michael@0:   printf("Test 1 - ToUpper(char16_t, char16_t*):\n");
michael@0:   for(i=0;i < T2LEN ; i++)
michael@0:   {
michael@0:     char16_t ch = ToUpperCase(t2data[i]);
michael@0:     if(ch != t2result[i])
michael@0:       printf("\tFailed!! result unexpected %d\n", i);
michael@0:   }
michael@0: 
michael@0: 
michael@0:   printf("Test 2 - ToLower(char16_t, char16_t*):\n");
michael@0:   for(i=0;i < T3LEN; i++)
michael@0:   {
michael@0:     char16_t ch = ToLowerCase(t3data[i]);
michael@0:     if(ch != t3result[i])
michael@0:       printf("\tFailed!! result unexpected %d\n", i);
michael@0:   }
michael@0: 
michael@0:   printf("Test 3 - ToTitle(char16_t, char16_t*):\n");
michael@0:   for(i=0;i < T4LEN; i++)
michael@0:   {
michael@0:     char16_t ch = ToTitleCase(t4data[i]);
michael@0:     if(ch != t4result[i])
michael@0:       printf("\tFailed!! result unexpected %d\n", i);
michael@0:   }
michael@0: 
michael@0:   printf("Test 4 - ToUpper(char16_t*, char16_t*, uint32_t):\n");
michael@0:   ToUpperCase(t2data, buf, T2LEN);
michael@0:   for(i = 0; i < T2LEN; i++)
michael@0:   {
michael@0:      if(buf[i] != t2result[i])
michael@0:      {
michael@0:        printf("\tFailed!! result unexpected %d\n", i);
michael@0:        break;
michael@0:      }
michael@0:   }
michael@0: 
michael@0:   printf("Test 5 - ToLower(char16_t*, char16_t*, uint32_t):\n");
michael@0:   ToLowerCase(t3data, buf, T3LEN);
michael@0:   for(i = 0; i < T3LEN; i++)
michael@0:   {
michael@0:      if(buf[i] != t3result[i])
michael@0:      {
michael@0:        printf("\tFailed!! result unexpected %d\n", i);
michael@0:        break;
michael@0:      }
michael@0:   }
michael@0: 
michael@0:   printf("Test 6 - CaseInsensitiveCompare UTF-8 (1):\n");
michael@0:   if (CaseInsensitiveCompare((char*)t6lhs, (char*)t6rhs, sizeof(t6lhs), sizeof(t6rhs)))
michael@0:     printf("\tFailed!\n");
michael@0:   if (!CharByCharCompareEqual((char*)t6lhs, (char*)t6rhs, sizeof(t6lhs), sizeof(t6rhs)))
michael@0:     printf("\tFailed character-by-character comparison!\n");
michael@0: 
michael@0:   printf("Test 7 - CaseInsensitiveCompare UTF-8 (2):\n");
michael@0:   if (CaseInsensitiveCompare(t7lhs, t7rhs, strlen(t7lhs), strlen(t7rhs)))
michael@0:     printf("\tFailed!\n");
michael@0:   if (!CharByCharCompareEqual(t7lhs, t7rhs, sizeof(t7lhs), sizeof(t7rhs)))
michael@0:     printf("\tFailed character-by-character comparison!\n");
michael@0: 
michael@0:   printf("Test 8a - CaseInsensitiveCompare UTF-8 (3):\n");
michael@0:   if (CaseInsensitiveCompare(t8lhs, t8rhs, strlen(t8lhs), strlen(t8rhs)) != -1)
michael@0:     printf("\tFailed!\n");
michael@0:   if (CharByCharCompareEqual(t8lhs, t8rhs, strlen(t8lhs), strlen(t8rhs)))
michael@0:     printf("\tFailed character-by-character comparison!\n");
michael@0: 
michael@0:   printf("Test 8b - CaseInsensitiveCompare UTF-8 (4):\n");
michael@0:   if (CaseInsensitiveCompare(t8rhs, t8lhs, strlen(t8rhs), strlen(t8lhs)) != 1)
michael@0:     printf("\tFailed!\n");
michael@0: 
michael@0:   // This test may seem a bit strange.  But it's actually an easy bug to make
michael@0:   // if we tried to be clever and say that two ASCII characters x and y are
michael@0:   // case-insensitively equal if (x & ~0x20) == (y & ~0x20).
michael@0:   printf("Test 9 - CaseInsensitiveCompare UTF-8 (5):\n");
michael@0:   if (CaseInsensitiveCompare(t9rhs, t9lhs, strlen(t9lhs), strlen(t9rhs)) != 1)
michael@0:     printf("\tFailed!\n");
michael@0:   if (CharByCharCompareEqual(t9lhs, t9rhs, strlen(t9lhs), strlen(t9rhs)))
michael@0:     printf("\tFailed character-by-character comparison!\n");
michael@0: 
michael@0:   printf("===========================\n");
michael@0:   printf("Finish case conversion test\n");
michael@0:   printf("===========================\n");
michael@0: }
michael@0: 
michael@0: static void FuzzOneInvalidCaseConversion()
michael@0: {
michael@0:   uint32_t aLen = rand() % 32;
michael@0:   uint32_t bLen = rand() % 32;
michael@0: 
michael@0:   // We could use a static length-32 buffer for these, but then Valgrind
michael@0:   // wouldn't be able to detect errors.
michael@0:   unsigned char *aBuf = (unsigned char*)malloc(aLen * sizeof(unsigned char));
michael@0:   unsigned char *bBuf = (unsigned char*)malloc(bLen * sizeof(unsigned char));
michael@0: 
michael@0:   for (uint32_t i = 0; i < aLen; i++) {
michael@0:     aBuf[i] = rand() & 0xff;
michael@0:   }
michael@0: 
michael@0:   for (uint32_t i = 0; i < bLen; i++) {
michael@0:     bBuf[i] = rand() & 0xff;
michael@0:   }
michael@0: 
michael@0:   if (!CaseInsensitiveCompare((char*)aBuf, (char*)bBuf, aLen, bLen))
michael@0:     printf("\tSurprise, two random strings compared insensitively as equal!\n");
michael@0:   if (CharByCharCompareEqual((char*)aBuf, (char*)bBuf, aLen, bLen))
michael@0:     printf("\tSurprise, two random strings compared as exactly equal!\n");
michael@0: 
michael@0:   free(aBuf);
michael@0:   free(bBuf);
michael@0: }
michael@0: 
michael@0: static void FuzzCaseConversion()
michael@0: {
michael@0:   printf("==========================\n");
michael@0:   printf("Start fuzz case conversion\n");
michael@0:   printf("==========================\n");
michael@0: 
michael@0:   srand(0);
michael@0: 
michael@0:   printf("Fuzzing invalid UTF8 data...\n");
michael@0:   for (uint32_t i = 0; i < 100000; i++) {
michael@0:     FuzzOneInvalidCaseConversion();
michael@0:   }
michael@0: 
michael@0:   printf("===========================\n");
michael@0:   printf("Finish fuzz case conversion\n");
michael@0:   printf("===========================\n");
michael@0: }
michael@0: 
michael@0: static void TestEntityConversion(uint32_t version)
michael@0: {
michael@0:   printf("==============================\n");
michael@0:   printf("Start nsIEntityConverter Test \n");
michael@0:   printf("==============================\n");
michael@0: 
michael@0:   uint32_t i;
michael@0:   nsString inString;
michael@0:   char16_t uChar;
michael@0:   nsresult res;
michael@0: 
michael@0: 
michael@0:   inString.Assign(NS_ConvertASCIItoUTF16("\xA0\xA1\xA2\xA3"));
michael@0:   uChar = (char16_t) 8364; //euro
michael@0:   inString.Append(&uChar, 1);
michael@0:   uChar = (char16_t) 9830; //
michael@0:   inString.Append(&uChar, 1);
michael@0: 
michael@0:   nsCOMPtr <nsIEntityConverter> entityConv = do_CreateInstance(kEntityConverterCID, &res);;
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n"); return;}
michael@0: 
michael@0:   const char16_t *data;
michael@0:   uint32_t length = NS_StringGetData(inString, &data);
michael@0: 
michael@0:   // convert char by char
michael@0:   for (i = 0; i < length; i++) {
michael@0:     char *entity = nullptr;
michael@0:     res = entityConv->ConvertToEntity(data[i], version, &entity);
michael@0:     if (NS_SUCCEEDED(res) && entity) {
michael@0:       printf("%c %s\n", data[i], entity);
michael@0:       nsMemory::Free(entity);
michael@0:     }
michael@0:   }
michael@0: 
michael@0:   // convert at once as a string
michael@0:   char16_t *entities;
michael@0:   res = entityConv->ConvertToEntities(inString.get(), version, &entities);
michael@0:   if (NS_SUCCEEDED(res) && entities) {
michael@0:     for (char16_t *centity = entities; *centity; ++centity) {
michael@0:       printf("%c", (char) *centity);
michael@0:       if (';' == (char) *centity)
michael@0:         printf("\n");
michael@0:     }
michael@0:     nsMemory::Free(entities);
michael@0:   }
michael@0: 
michael@0:   printf("==============================\n");
michael@0:   printf("Finish nsIEntityConverter Test \n");
michael@0:   printf("==============================\n\n");
michael@0: }
michael@0: 
michael@0: static void TestSaveAsCharset()
michael@0: {
michael@0:   printf("==============================\n");
michael@0:   printf("Start nsISaveAsCharset Test \n");
michael@0:   printf("==============================\n");
michael@0: 
michael@0:   nsresult res;
michael@0: 
michael@0:   nsString inString;
michael@0:   inString.Assign(NS_ConvertASCIItoUTF16("\x61\x62\x80\xA0\x63"));
michael@0:   char *outString;
michael@0:   
michael@0:   const char16_t *data;
michael@0:   uint32_t length = NS_StringGetData(inString, &data);
michael@0: 
michael@0:   // first, dump input string
michael@0:   for (uint32_t i = 0; i < length; i++) {
michael@0:     printf("%c ", data[i]);
michael@0:   }
michael@0:   printf("\n");
michael@0: 
michael@0:   nsCOMPtr <nsISaveAsCharset> saveAsCharset = do_CreateInstance(kSaveAsCharsetCID, &res);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   
michael@0:   printf("ISO-8859-1 attr_plainTextDefault entityNone\n");
michael@0:   res = saveAsCharset->Init("ISO-8859-1", 
michael@0:                                  nsISaveAsCharset::attr_plainTextDefault, 
michael@0:                                  nsIEntityConverter::entityNone);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0: 
michael@0:   printf("ISO-2022-JP attr_plainTextDefault entityNone\n");
michael@0:   res = saveAsCharset->Init("ISO-2022-JP", 
michael@0:                                  nsISaveAsCharset::attr_plainTextDefault,
michael@0:                                  nsIEntityConverter::entityNone);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0:   if (NS_ERROR_UENC_NOMAPPING == res) {
michael@0:     outString = ToNewUTF8String(inString);
michael@0:     if (!outString) {printf("\tFailed!! output null\n");}
michael@0:     else {printf("Fall back to UTF-8: %s\n", outString); nsMemory::Free(outString);}
michael@0:   }
michael@0: 
michael@0:   printf("ISO-2022-JP attr_FallbackQuestionMark entityNone\n");
michael@0:   res = saveAsCharset->Init("ISO-2022-JP", 
michael@0:                                  nsISaveAsCharset::attr_FallbackQuestionMark,
michael@0:                                  nsIEntityConverter::entityNone);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0: 
michael@0:   printf("ISO-2022-JP attr_FallbackEscapeU entityNone\n");
michael@0:   res = saveAsCharset->Init("ISO-2022-JP", 
michael@0:                                  nsISaveAsCharset::attr_FallbackEscapeU,
michael@0:                                  nsIEntityConverter::entityNone);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0: 
michael@0:   printf("ISO-8859-1 attr_htmlTextDefault html40Latin1\n");
michael@0:   res = saveAsCharset->Init("ISO-8859-1", 
michael@0:                                  nsISaveAsCharset::attr_htmlTextDefault, 
michael@0:                                  nsIEntityConverter::html40Latin1);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_ERROR_UENC_NOMAPPING != res && NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0: 
michael@0:   printf("ISO-8859-1 attr_FallbackHexNCR+attr_EntityAfterCharsetConv html40Latin1 \n");
michael@0:   res = saveAsCharset->Init("ISO-8859-1", 
michael@0:                                  nsISaveAsCharset::attr_FallbackHexNCR + 
michael@0:                                  nsISaveAsCharset::attr_EntityAfterCharsetConv, 
michael@0:                                  nsIEntityConverter::html40Latin1);
michael@0:   if (NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   res = saveAsCharset->Convert(inString.get(), &outString);
michael@0:   if (NS_ERROR_UENC_NOMAPPING != res && NS_FAILED(res)) {printf("\tFailed!! return value != NS_OK\n");}
michael@0:   if (!outString) {printf("\tFailed!! output null\n");}
michael@0:   else {printf("%s\n", outString); nsMemory::Free(outString);}
michael@0: 
michael@0: 
michael@0:   printf("==============================\n");
michael@0:   printf("Finish nsISaveAsCharset Test \n");
michael@0:   printf("==============================\n\n");
michael@0: }
michael@0: 
michael@0: static char16_t normStr[] = 
michael@0: {
michael@0:   0x00E1,   
michael@0:   0x0061,
michael@0:   0x0301,
michael@0:   0x0107,
michael@0:   0x0063,
michael@0:   0x0301,
michael@0:   0x0000
michael@0: };
michael@0: 
michael@0: static char16_t nfdForm[] = 
michael@0: {
michael@0:   0x0061,
michael@0:   0x0301,
michael@0:   0x0061,
michael@0:   0x0301,
michael@0:   0x0063,
michael@0:   0x0301,
michael@0:   0x0063,
michael@0:   0x0301,
michael@0:   0x0000
michael@0: };
michael@0: 
michael@0: void TestNormalization()
michael@0: {
michael@0:    printf("==============================\n");
michael@0:    printf("Start nsIUnicodeNormalizer Test \n");
michael@0:    printf("==============================\n");
michael@0:    nsIUnicodeNormalizer *t = nullptr;
michael@0:    nsresult res;
michael@0:    res = CallGetService(kUnicodeNormalizerCID, &t);
michael@0:            
michael@0:    printf("Test 1 - GetService():\n");
michael@0:    if(NS_FAILED(res) || !t) {
michael@0:      printf("\t1st Norm GetService failed\n");
michael@0:    } else {
michael@0:      NS_RELEASE(t);
michael@0:    }
michael@0: 
michael@0:    res = CallGetService(kUnicodeNormalizerCID, &t);
michael@0:            
michael@0:    if(NS_FAILED(res) || !t) {
michael@0:      printf("\t2nd GetService failed\n");
michael@0:    } else {
michael@0:     printf("Test 2 - NormalizeUnicode(uint32_t, const nsAString&, nsAString&):\n");
michael@0:     nsAutoString resultStr;
michael@0:     res =  t->NormalizeUnicodeNFD(nsDependentString(normStr), resultStr);
michael@0:     if (resultStr.Equals(nsDependentString(nfdForm))) {
michael@0:       printf(" Succeeded in NFD UnicodeNormalizer test. \n");
michael@0:     } else {
michael@0:       printf(" Failed in NFD UnicodeNormalizer test. \n");
michael@0:     }
michael@0: 
michael@0:     NS_RELEASE(t);
michael@0:    }
michael@0:    printf("==============================\n");
michael@0:    printf("Finish nsIUnicodeNormalizer Test \n");
michael@0:    printf("==============================\n");
michael@0: 
michael@0: }
michael@0: 
michael@0: 
michael@0: int main(int argc, char** argv) {
michael@0:    
michael@0:    nsresult rv = NS_InitXPCOM2(nullptr, nullptr, nullptr);
michael@0:    if (NS_FAILED(rv)) {
michael@0:       printf("NS_InitXPCOM2 failed\n");
michael@0:       return 1;
michael@0:    }
michael@0: 
michael@0:    // --------------------------------------------
michael@0: 
michael@0:    TestCaseConversion();
michael@0: 
michael@0:    // --------------------------------------------
michael@0: 
michael@0:    FuzzCaseConversion();
michael@0: 
michael@0:    // --------------------------------------------
michael@0: 
michael@0:    TestEntityConversion(nsIEntityConverter::html40);
michael@0: 
michael@0:    // --------------------------------------------
michael@0: 
michael@0:    TestSaveAsCharset();
michael@0: 
michael@0:    // --------------------------------------------
michael@0: 
michael@0:    TestNormalization();
michael@0: 
michael@0:    // --------------------------------------------
michael@0:    printf("Finish All The Test Cases\n");
michael@0: 
michael@0:    return 0;
michael@0: }