intl/lwbrk/nsSampleWordBreaker.cpp

   1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
   2 /* This Source Code Form is subject to the terms of the Mozilla Public
   3  * License, v. 2.0. If a copy of the MPL was not distributed with this
   4  * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
   5
   6
   7 #include "nsSampleWordBreaker.h"
   8
   9 nsSampleWordBreaker::nsSampleWordBreaker()
  10 {
  11 }
  12 nsSampleWordBreaker::~nsSampleWordBreaker()
  13 {
  14 }
  15
  16 NS_IMPL_ISUPPORTS(nsSampleWordBreaker, nsIWordBreaker)
  17
  18 bool nsSampleWordBreaker::BreakInBetween(
  19   const char16_t* aText1 , uint32_t aTextLen1,
  20   const char16_t* aText2 , uint32_t aTextLen2)
  21 {
  22   NS_PRECONDITION( nullptr != aText1, "null ptr");
  23   NS_PRECONDITION( nullptr != aText2, "null ptr");
  24
  25   if(!aText1 || !aText2 || (0 == aTextLen1) || (0 == aTextLen2))
  26     return false;
  27
  28   return (this->GetClass(aText1[aTextLen1-1]) != this->GetClass(aText2[0]));
  29 }
  30
  31
  32 #define IS_ASCII(c)            (0 == ( 0xFF80 & (c)))
  33 #define ASCII_IS_ALPHA(c)         ((( 'a' <= (c)) && ((c) <= 'z')) || (( 'A' <= (c)) && ((c) <= 'Z')))
  34 #define ASCII_IS_DIGIT(c)         (( '0' <= (c)) && ((c) <= '9'))
  35 #define ASCII_IS_SPACE(c)         (( ' ' == (c)) || ( '\t' == (c)) || ( '\r' == (c)) || ( '\n' == (c)))
  36 #define IS_ALPHABETICAL_SCRIPT(c) ((c) < 0x2E80)
  37
  38 // we change the beginning of IS_HAN from 0x4e00 to 0x3400 to relfect Unicode 3.0
  39 #define IS_HAN(c)              (( 0x3400 <= (c)) && ((c) <= 0x9fff))||(( 0xf900 <= (c)) && ((c) <= 0xfaff))
  40 #define IS_KATAKANA(c)         (( 0x30A0 <= (c)) && ((c) <= 0x30FF))
  41 #define IS_HIRAGANA(c)         (( 0x3040 <= (c)) && ((c) <= 0x309F))
  42 #define IS_HALFWIDTHKATAKANA(c)         (( 0xFF60 <= (c)) && ((c) <= 0xFF9F))
  43 #define IS_THAI(c)         (0x0E00 == (0xFF80 & (c) )) // Look at the higest 9 bits
  44
  45 uint8_t nsSampleWordBreaker::GetClass(char16_t c)
  46 {
  47   // begin of the hack
  48
  49   if (IS_ALPHABETICAL_SCRIPT(c))  {
  50           if(IS_ASCII(c))  {
  51                   if(ASCII_IS_SPACE(c)) {
  52                           return kWbClassSpace;
  53                   } else if(ASCII_IS_ALPHA(c) || ASCII_IS_DIGIT(c)) {
  54                           return kWbClassAlphaLetter;
  55                   } else {
  56                           return kWbClassPunct;
  57                   }
  58           } else if(IS_THAI(c)) {
  59                   return kWbClassThaiLetter;
  60           } else if (c == 0x00A0/*NBSP*/) {
  61       return kWbClassSpace;
  62     } else {
  63                   return kWbClassAlphaLetter;
  64           }
  65   }  else {
  66           if(IS_HAN(c)) {
  67                   return kWbClassHanLetter;
  68           } else if(IS_KATAKANA(c))   {
  69                   return kWbClassKatakanaLetter;
  70           } else if(IS_HIRAGANA(c))   {
  71                   return kWbClassHiraganaLetter;
  72           } else if(IS_HALFWIDTHKATAKANA(c))  {
  73                   return kWbClassHWKatakanaLetter;
  74           } else  {
  75                   return kWbClassAlphaLetter;
  76           }
  77   }
  78   return 0;
  79 }
  80
  81 nsWordRange nsSampleWordBreaker::FindWord(
  82   const char16_t* aText , uint32_t aTextLen,
  83   uint32_t aOffset)
  84 {
  85   nsWordRange range;
  86   NS_PRECONDITION( nullptr != aText, "null ptr");
  87   NS_PRECONDITION( 0 != aTextLen, "len = 0");
  88   NS_PRECONDITION( aOffset <= aTextLen, "aOffset > aTextLen");
  89
  90   range.mBegin = aTextLen + 1;
  91   range.mEnd = aTextLen + 1;
  92
  93   if(!aText || aOffset > aTextLen)
  94     return range;
  95
  96   uint8_t c = this->GetClass(aText[aOffset]);
  97   uint32_t i;
  98   // Scan forward
  99   range.mEnd--;
 100   for(i = aOffset +1;i <= aTextLen; i++)
 101   {
 102      if( c != this->GetClass(aText[i]))
 103      {
 104        range.mEnd = i;
 105        break;
 106      }
 107   }
 108
 109   // Scan backward
 110   range.mBegin = 0;
 111   for(i = aOffset ;i > 0; i--)
 112   {
 113      if( c != this->GetClass(aText[i-1]))
 114      {
 115        range.mBegin = i;
 116        break;
 117      }
 118   }
 119   if(kWbClassThaiLetter == c)
 120   {
 121         // need to call Thai word breaker from here
 122         // we should pass the whole Thai segment to the thai word breaker to find a shorter answer
 123   }
 124   return range;
 125 }
 126
 127 int32_t nsSampleWordBreaker::NextWord(
 128   const char16_t* aText, uint32_t aLen, uint32_t aPos)
 129 {
 130   int8_t c1, c2;
 131   uint32_t cur = aPos;
 132   if (cur == aLen)
 133     return NS_WORDBREAKER_NEED_MORE_TEXT;
 134   c1 = this->GetClass(aText[cur]);
 135
 136   for(cur++; cur <aLen; cur++)
 137   {
 138      c2 = this->GetClass(aText[cur]);
 139      if(c2 != c1)
 140        break;
 141   }
 142   if(kWbClassThaiLetter == c1)
 143   {
 144         // need to call Thai word breaker from here
 145         // we should pass the whole Thai segment to the thai word breaker to find a shorter answer
 146   }
 147   if (cur == aLen)
 148     return NS_WORDBREAKER_NEED_MORE_TEXT;
 149   return cur;
 150 }