Wed, 31 Dec 2014 06:09:35 +0100
Cloned upstream origin tor-browser at tor-browser-31.3.0esr-4.5-1-build1
revision ID fc1c9ff7c1b2defdbc039f12214767608f46423f for hacking purpose.
michael@0 | 1 | /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ |
michael@0 | 2 | /* This Source Code Form is subject to the terms of the Mozilla Public |
michael@0 | 3 | * License, v. 2.0. If a copy of the MPL was not distributed with this |
michael@0 | 4 | * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ |
michael@0 | 5 | |
michael@0 | 6 | |
michael@0 | 7 | #include "nsSampleWordBreaker.h" |
michael@0 | 8 | |
michael@0 | 9 | nsSampleWordBreaker::nsSampleWordBreaker() |
michael@0 | 10 | { |
michael@0 | 11 | } |
michael@0 | 12 | nsSampleWordBreaker::~nsSampleWordBreaker() |
michael@0 | 13 | { |
michael@0 | 14 | } |
michael@0 | 15 | |
michael@0 | 16 | NS_IMPL_ISUPPORTS(nsSampleWordBreaker, nsIWordBreaker) |
michael@0 | 17 | |
michael@0 | 18 | bool nsSampleWordBreaker::BreakInBetween( |
michael@0 | 19 | const char16_t* aText1 , uint32_t aTextLen1, |
michael@0 | 20 | const char16_t* aText2 , uint32_t aTextLen2) |
michael@0 | 21 | { |
michael@0 | 22 | NS_PRECONDITION( nullptr != aText1, "null ptr"); |
michael@0 | 23 | NS_PRECONDITION( nullptr != aText2, "null ptr"); |
michael@0 | 24 | |
michael@0 | 25 | if(!aText1 || !aText2 || (0 == aTextLen1) || (0 == aTextLen2)) |
michael@0 | 26 | return false; |
michael@0 | 27 | |
michael@0 | 28 | return (this->GetClass(aText1[aTextLen1-1]) != this->GetClass(aText2[0])); |
michael@0 | 29 | } |
michael@0 | 30 | |
michael@0 | 31 | |
michael@0 | 32 | #define IS_ASCII(c) (0 == ( 0xFF80 & (c))) |
michael@0 | 33 | #define ASCII_IS_ALPHA(c) ((( 'a' <= (c)) && ((c) <= 'z')) || (( 'A' <= (c)) && ((c) <= 'Z'))) |
michael@0 | 34 | #define ASCII_IS_DIGIT(c) (( '0' <= (c)) && ((c) <= '9')) |
michael@0 | 35 | #define ASCII_IS_SPACE(c) (( ' ' == (c)) || ( '\t' == (c)) || ( '\r' == (c)) || ( '\n' == (c))) |
michael@0 | 36 | #define IS_ALPHABETICAL_SCRIPT(c) ((c) < 0x2E80) |
michael@0 | 37 | |
michael@0 | 38 | // we change the beginning of IS_HAN from 0x4e00 to 0x3400 to relfect Unicode 3.0 |
michael@0 | 39 | #define IS_HAN(c) (( 0x3400 <= (c)) && ((c) <= 0x9fff))||(( 0xf900 <= (c)) && ((c) <= 0xfaff)) |
michael@0 | 40 | #define IS_KATAKANA(c) (( 0x30A0 <= (c)) && ((c) <= 0x30FF)) |
michael@0 | 41 | #define IS_HIRAGANA(c) (( 0x3040 <= (c)) && ((c) <= 0x309F)) |
michael@0 | 42 | #define IS_HALFWIDTHKATAKANA(c) (( 0xFF60 <= (c)) && ((c) <= 0xFF9F)) |
michael@0 | 43 | #define IS_THAI(c) (0x0E00 == (0xFF80 & (c) )) // Look at the higest 9 bits |
michael@0 | 44 | |
michael@0 | 45 | uint8_t nsSampleWordBreaker::GetClass(char16_t c) |
michael@0 | 46 | { |
michael@0 | 47 | // begin of the hack |
michael@0 | 48 | |
michael@0 | 49 | if (IS_ALPHABETICAL_SCRIPT(c)) { |
michael@0 | 50 | if(IS_ASCII(c)) { |
michael@0 | 51 | if(ASCII_IS_SPACE(c)) { |
michael@0 | 52 | return kWbClassSpace; |
michael@0 | 53 | } else if(ASCII_IS_ALPHA(c) || ASCII_IS_DIGIT(c)) { |
michael@0 | 54 | return kWbClassAlphaLetter; |
michael@0 | 55 | } else { |
michael@0 | 56 | return kWbClassPunct; |
michael@0 | 57 | } |
michael@0 | 58 | } else if(IS_THAI(c)) { |
michael@0 | 59 | return kWbClassThaiLetter; |
michael@0 | 60 | } else if (c == 0x00A0/*NBSP*/) { |
michael@0 | 61 | return kWbClassSpace; |
michael@0 | 62 | } else { |
michael@0 | 63 | return kWbClassAlphaLetter; |
michael@0 | 64 | } |
michael@0 | 65 | } else { |
michael@0 | 66 | if(IS_HAN(c)) { |
michael@0 | 67 | return kWbClassHanLetter; |
michael@0 | 68 | } else if(IS_KATAKANA(c)) { |
michael@0 | 69 | return kWbClassKatakanaLetter; |
michael@0 | 70 | } else if(IS_HIRAGANA(c)) { |
michael@0 | 71 | return kWbClassHiraganaLetter; |
michael@0 | 72 | } else if(IS_HALFWIDTHKATAKANA(c)) { |
michael@0 | 73 | return kWbClassHWKatakanaLetter; |
michael@0 | 74 | } else { |
michael@0 | 75 | return kWbClassAlphaLetter; |
michael@0 | 76 | } |
michael@0 | 77 | } |
michael@0 | 78 | return 0; |
michael@0 | 79 | } |
michael@0 | 80 | |
michael@0 | 81 | nsWordRange nsSampleWordBreaker::FindWord( |
michael@0 | 82 | const char16_t* aText , uint32_t aTextLen, |
michael@0 | 83 | uint32_t aOffset) |
michael@0 | 84 | { |
michael@0 | 85 | nsWordRange range; |
michael@0 | 86 | NS_PRECONDITION( nullptr != aText, "null ptr"); |
michael@0 | 87 | NS_PRECONDITION( 0 != aTextLen, "len = 0"); |
michael@0 | 88 | NS_PRECONDITION( aOffset <= aTextLen, "aOffset > aTextLen"); |
michael@0 | 89 | |
michael@0 | 90 | range.mBegin = aTextLen + 1; |
michael@0 | 91 | range.mEnd = aTextLen + 1; |
michael@0 | 92 | |
michael@0 | 93 | if(!aText || aOffset > aTextLen) |
michael@0 | 94 | return range; |
michael@0 | 95 | |
michael@0 | 96 | uint8_t c = this->GetClass(aText[aOffset]); |
michael@0 | 97 | uint32_t i; |
michael@0 | 98 | // Scan forward |
michael@0 | 99 | range.mEnd--; |
michael@0 | 100 | for(i = aOffset +1;i <= aTextLen; i++) |
michael@0 | 101 | { |
michael@0 | 102 | if( c != this->GetClass(aText[i])) |
michael@0 | 103 | { |
michael@0 | 104 | range.mEnd = i; |
michael@0 | 105 | break; |
michael@0 | 106 | } |
michael@0 | 107 | } |
michael@0 | 108 | |
michael@0 | 109 | // Scan backward |
michael@0 | 110 | range.mBegin = 0; |
michael@0 | 111 | for(i = aOffset ;i > 0; i--) |
michael@0 | 112 | { |
michael@0 | 113 | if( c != this->GetClass(aText[i-1])) |
michael@0 | 114 | { |
michael@0 | 115 | range.mBegin = i; |
michael@0 | 116 | break; |
michael@0 | 117 | } |
michael@0 | 118 | } |
michael@0 | 119 | if(kWbClassThaiLetter == c) |
michael@0 | 120 | { |
michael@0 | 121 | // need to call Thai word breaker from here |
michael@0 | 122 | // we should pass the whole Thai segment to the thai word breaker to find a shorter answer |
michael@0 | 123 | } |
michael@0 | 124 | return range; |
michael@0 | 125 | } |
michael@0 | 126 | |
michael@0 | 127 | int32_t nsSampleWordBreaker::NextWord( |
michael@0 | 128 | const char16_t* aText, uint32_t aLen, uint32_t aPos) |
michael@0 | 129 | { |
michael@0 | 130 | int8_t c1, c2; |
michael@0 | 131 | uint32_t cur = aPos; |
michael@0 | 132 | if (cur == aLen) |
michael@0 | 133 | return NS_WORDBREAKER_NEED_MORE_TEXT; |
michael@0 | 134 | c1 = this->GetClass(aText[cur]); |
michael@0 | 135 | |
michael@0 | 136 | for(cur++; cur <aLen; cur++) |
michael@0 | 137 | { |
michael@0 | 138 | c2 = this->GetClass(aText[cur]); |
michael@0 | 139 | if(c2 != c1) |
michael@0 | 140 | break; |
michael@0 | 141 | } |
michael@0 | 142 | if(kWbClassThaiLetter == c1) |
michael@0 | 143 | { |
michael@0 | 144 | // need to call Thai word breaker from here |
michael@0 | 145 | // we should pass the whole Thai segment to the thai word breaker to find a shorter answer |
michael@0 | 146 | } |
michael@0 | 147 | if (cur == aLen) |
michael@0 | 148 | return NS_WORDBREAKER_NEED_MORE_TEXT; |
michael@0 | 149 | return cur; |
michael@0 | 150 | } |