1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/lwbrk/src/nsSampleWordBreaker.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,150 @@
1.4 +/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +
1.10 +#include "nsSampleWordBreaker.h"
1.11 +
1.12 +nsSampleWordBreaker::nsSampleWordBreaker()
1.13 +{
1.14 +}
1.15 +nsSampleWordBreaker::~nsSampleWordBreaker()
1.16 +{
1.17 +}
1.18 +
1.19 +NS_IMPL_ISUPPORTS(nsSampleWordBreaker, nsIWordBreaker)
1.20 +
1.21 +bool nsSampleWordBreaker::BreakInBetween(
1.22 + const char16_t* aText1 , uint32_t aTextLen1,
1.23 + const char16_t* aText2 , uint32_t aTextLen2)
1.24 +{
1.25 + NS_PRECONDITION( nullptr != aText1, "null ptr");
1.26 + NS_PRECONDITION( nullptr != aText2, "null ptr");
1.27 +
1.28 + if(!aText1 || !aText2 || (0 == aTextLen1) || (0 == aTextLen2))
1.29 + return false;
1.30 +
1.31 + return (this->GetClass(aText1[aTextLen1-1]) != this->GetClass(aText2[0]));
1.32 +}
1.33 +
1.34 +
1.35 +#define IS_ASCII(c) (0 == ( 0xFF80 & (c)))
1.36 +#define ASCII_IS_ALPHA(c) ((( 'a' <= (c)) && ((c) <= 'z')) || (( 'A' <= (c)) && ((c) <= 'Z')))
1.37 +#define ASCII_IS_DIGIT(c) (( '0' <= (c)) && ((c) <= '9'))
1.38 +#define ASCII_IS_SPACE(c) (( ' ' == (c)) || ( '\t' == (c)) || ( '\r' == (c)) || ( '\n' == (c)))
1.39 +#define IS_ALPHABETICAL_SCRIPT(c) ((c) < 0x2E80)
1.40 +
1.41 +// we change the beginning of IS_HAN from 0x4e00 to 0x3400 to relfect Unicode 3.0
1.42 +#define IS_HAN(c) (( 0x3400 <= (c)) && ((c) <= 0x9fff))||(( 0xf900 <= (c)) && ((c) <= 0xfaff))
1.43 +#define IS_KATAKANA(c) (( 0x30A0 <= (c)) && ((c) <= 0x30FF))
1.44 +#define IS_HIRAGANA(c) (( 0x3040 <= (c)) && ((c) <= 0x309F))
1.45 +#define IS_HALFWIDTHKATAKANA(c) (( 0xFF60 <= (c)) && ((c) <= 0xFF9F))
1.46 +#define IS_THAI(c) (0x0E00 == (0xFF80 & (c) )) // Look at the higest 9 bits
1.47 +
1.48 +uint8_t nsSampleWordBreaker::GetClass(char16_t c)
1.49 +{
1.50 + // begin of the hack
1.51 +
1.52 + if (IS_ALPHABETICAL_SCRIPT(c)) {
1.53 + if(IS_ASCII(c)) {
1.54 + if(ASCII_IS_SPACE(c)) {
1.55 + return kWbClassSpace;
1.56 + } else if(ASCII_IS_ALPHA(c) || ASCII_IS_DIGIT(c)) {
1.57 + return kWbClassAlphaLetter;
1.58 + } else {
1.59 + return kWbClassPunct;
1.60 + }
1.61 + } else if(IS_THAI(c)) {
1.62 + return kWbClassThaiLetter;
1.63 + } else if (c == 0x00A0/*NBSP*/) {
1.64 + return kWbClassSpace;
1.65 + } else {
1.66 + return kWbClassAlphaLetter;
1.67 + }
1.68 + } else {
1.69 + if(IS_HAN(c)) {
1.70 + return kWbClassHanLetter;
1.71 + } else if(IS_KATAKANA(c)) {
1.72 + return kWbClassKatakanaLetter;
1.73 + } else if(IS_HIRAGANA(c)) {
1.74 + return kWbClassHiraganaLetter;
1.75 + } else if(IS_HALFWIDTHKATAKANA(c)) {
1.76 + return kWbClassHWKatakanaLetter;
1.77 + } else {
1.78 + return kWbClassAlphaLetter;
1.79 + }
1.80 + }
1.81 + return 0;
1.82 +}
1.83 +
1.84 +nsWordRange nsSampleWordBreaker::FindWord(
1.85 + const char16_t* aText , uint32_t aTextLen,
1.86 + uint32_t aOffset)
1.87 +{
1.88 + nsWordRange range;
1.89 + NS_PRECONDITION( nullptr != aText, "null ptr");
1.90 + NS_PRECONDITION( 0 != aTextLen, "len = 0");
1.91 + NS_PRECONDITION( aOffset <= aTextLen, "aOffset > aTextLen");
1.92 +
1.93 + range.mBegin = aTextLen + 1;
1.94 + range.mEnd = aTextLen + 1;
1.95 +
1.96 + if(!aText || aOffset > aTextLen)
1.97 + return range;
1.98 +
1.99 + uint8_t c = this->GetClass(aText[aOffset]);
1.100 + uint32_t i;
1.101 + // Scan forward
1.102 + range.mEnd--;
1.103 + for(i = aOffset +1;i <= aTextLen; i++)
1.104 + {
1.105 + if( c != this->GetClass(aText[i]))
1.106 + {
1.107 + range.mEnd = i;
1.108 + break;
1.109 + }
1.110 + }
1.111 +
1.112 + // Scan backward
1.113 + range.mBegin = 0;
1.114 + for(i = aOffset ;i > 0; i--)
1.115 + {
1.116 + if( c != this->GetClass(aText[i-1]))
1.117 + {
1.118 + range.mBegin = i;
1.119 + break;
1.120 + }
1.121 + }
1.122 + if(kWbClassThaiLetter == c)
1.123 + {
1.124 + // need to call Thai word breaker from here
1.125 + // we should pass the whole Thai segment to the thai word breaker to find a shorter answer
1.126 + }
1.127 + return range;
1.128 +}
1.129 +
1.130 +int32_t nsSampleWordBreaker::NextWord(
1.131 + const char16_t* aText, uint32_t aLen, uint32_t aPos)
1.132 +{
1.133 + int8_t c1, c2;
1.134 + uint32_t cur = aPos;
1.135 + if (cur == aLen)
1.136 + return NS_WORDBREAKER_NEED_MORE_TEXT;
1.137 + c1 = this->GetClass(aText[cur]);
1.138 +
1.139 + for(cur++; cur <aLen; cur++)
1.140 + {
1.141 + c2 = this->GetClass(aText[cur]);
1.142 + if(c2 != c1)
1.143 + break;
1.144 + }
1.145 + if(kWbClassThaiLetter == c1)
1.146 + {
1.147 + // need to call Thai word breaker from here
1.148 + // we should pass the whole Thai segment to the thai word breaker to find a shorter answer
1.149 + }
1.150 + if (cur == aLen)
1.151 + return NS_WORDBREAKER_NEED_MORE_TEXT;
1.152 + return cur;
1.153 +}
