gecko-dev/intl/lwbrk/nsSampleWordBreaker.cpp
Birunthan Mohanathas 790694c449 Bug 1038537 - Part 5: Flatten intl/lwbrk/{idl,public,src}/ directories. r=smontagu
--HG--
rename : intl/lwbrk/src/Makefile.in => intl/lwbrk/Makefile.in
rename : intl/lwbrk/src/crashtests/416721.html => intl/lwbrk/crashtests/416721.html
rename : intl/lwbrk/src/crashtests/crashtests.list => intl/lwbrk/crashtests/crashtests.list
rename : intl/lwbrk/src/jisx4051class.h => intl/lwbrk/jisx4051class.h
rename : intl/lwbrk/src/jisx4051pairtable.txt => intl/lwbrk/jisx4051pairtable.txt
rename : intl/lwbrk/src/nsCarbonBreaker.cpp => intl/lwbrk/nsCarbonBreaker.cpp
rename : intl/lwbrk/src/nsComplexBreaker.h => intl/lwbrk/nsComplexBreaker.h
rename : intl/lwbrk/public/nsILineBreaker.h => intl/lwbrk/nsILineBreaker.h
rename : intl/lwbrk/idl/nsISemanticUnitScanner.idl => intl/lwbrk/nsISemanticUnitScanner.idl
rename : intl/lwbrk/public/nsIWordBreaker.h => intl/lwbrk/nsIWordBreaker.h
rename : intl/lwbrk/src/nsJISx4051LineBreaker.cpp => intl/lwbrk/nsJISx4051LineBreaker.cpp
rename : intl/lwbrk/src/nsJISx4051LineBreaker.h => intl/lwbrk/nsJISx4051LineBreaker.h
rename : intl/lwbrk/public/nsLWBrkCIID.h => intl/lwbrk/nsLWBrkCIID.h
rename : intl/lwbrk/src/nsPangoBreaker.cpp => intl/lwbrk/nsPangoBreaker.cpp
rename : intl/lwbrk/src/nsRuleBreaker.cpp => intl/lwbrk/nsRuleBreaker.cpp
rename : intl/lwbrk/src/nsSampleWordBreaker.cpp => intl/lwbrk/nsSampleWordBreaker.cpp
rename : intl/lwbrk/src/nsSampleWordBreaker.h => intl/lwbrk/nsSampleWordBreaker.h
rename : intl/lwbrk/src/nsSemanticUnitScanner.cpp => intl/lwbrk/nsSemanticUnitScanner.cpp
rename : intl/lwbrk/src/nsSemanticUnitScanner.h => intl/lwbrk/nsSemanticUnitScanner.h
rename : intl/lwbrk/src/nsUniscribeBreaker.cpp => intl/lwbrk/nsUniscribeBreaker.cpp
rename : intl/lwbrk/src/rulebrk.c => intl/lwbrk/rulebrk.c
rename : intl/lwbrk/src/rulebrk.h => intl/lwbrk/rulebrk.h
rename : intl/lwbrk/src/th_char.h => intl/lwbrk/th_char.h
2014-07-24 10:56:39 -07:00

151 lines
4.1 KiB
C++

/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "nsSampleWordBreaker.h"
nsSampleWordBreaker::nsSampleWordBreaker()
{
}
nsSampleWordBreaker::~nsSampleWordBreaker()
{
}
NS_IMPL_ISUPPORTS(nsSampleWordBreaker, nsIWordBreaker)
bool nsSampleWordBreaker::BreakInBetween(
const char16_t* aText1 , uint32_t aTextLen1,
const char16_t* aText2 , uint32_t aTextLen2)
{
NS_PRECONDITION( nullptr != aText1, "null ptr");
NS_PRECONDITION( nullptr != aText2, "null ptr");
if(!aText1 || !aText2 || (0 == aTextLen1) || (0 == aTextLen2))
return false;
return (this->GetClass(aText1[aTextLen1-1]) != this->GetClass(aText2[0]));
}
#define IS_ASCII(c) (0 == ( 0xFF80 & (c)))
#define ASCII_IS_ALPHA(c) ((( 'a' <= (c)) && ((c) <= 'z')) || (( 'A' <= (c)) && ((c) <= 'Z')))
#define ASCII_IS_DIGIT(c) (( '0' <= (c)) && ((c) <= '9'))
#define ASCII_IS_SPACE(c) (( ' ' == (c)) || ( '\t' == (c)) || ( '\r' == (c)) || ( '\n' == (c)))
#define IS_ALPHABETICAL_SCRIPT(c) ((c) < 0x2E80)
// we change the beginning of IS_HAN from 0x4e00 to 0x3400 to relfect Unicode 3.0
#define IS_HAN(c) (( 0x3400 <= (c)) && ((c) <= 0x9fff))||(( 0xf900 <= (c)) && ((c) <= 0xfaff))
#define IS_KATAKANA(c) (( 0x30A0 <= (c)) && ((c) <= 0x30FF))
#define IS_HIRAGANA(c) (( 0x3040 <= (c)) && ((c) <= 0x309F))
#define IS_HALFWIDTHKATAKANA(c) (( 0xFF60 <= (c)) && ((c) <= 0xFF9F))
#define IS_THAI(c) (0x0E00 == (0xFF80 & (c) )) // Look at the higest 9 bits
uint8_t nsSampleWordBreaker::GetClass(char16_t c)
{
// begin of the hack
if (IS_ALPHABETICAL_SCRIPT(c)) {
if(IS_ASCII(c)) {
if(ASCII_IS_SPACE(c)) {
return kWbClassSpace;
} else if(ASCII_IS_ALPHA(c) || ASCII_IS_DIGIT(c)) {
return kWbClassAlphaLetter;
} else {
return kWbClassPunct;
}
} else if(IS_THAI(c)) {
return kWbClassThaiLetter;
} else if (c == 0x00A0/*NBSP*/) {
return kWbClassSpace;
} else {
return kWbClassAlphaLetter;
}
} else {
if(IS_HAN(c)) {
return kWbClassHanLetter;
} else if(IS_KATAKANA(c)) {
return kWbClassKatakanaLetter;
} else if(IS_HIRAGANA(c)) {
return kWbClassHiraganaLetter;
} else if(IS_HALFWIDTHKATAKANA(c)) {
return kWbClassHWKatakanaLetter;
} else {
return kWbClassAlphaLetter;
}
}
return 0;
}
nsWordRange nsSampleWordBreaker::FindWord(
const char16_t* aText , uint32_t aTextLen,
uint32_t aOffset)
{
nsWordRange range;
NS_PRECONDITION( nullptr != aText, "null ptr");
NS_PRECONDITION( 0 != aTextLen, "len = 0");
NS_PRECONDITION( aOffset <= aTextLen, "aOffset > aTextLen");
range.mBegin = aTextLen + 1;
range.mEnd = aTextLen + 1;
if(!aText || aOffset > aTextLen)
return range;
uint8_t c = this->GetClass(aText[aOffset]);
uint32_t i;
// Scan forward
range.mEnd--;
for(i = aOffset +1;i <= aTextLen; i++)
{
if( c != this->GetClass(aText[i]))
{
range.mEnd = i;
break;
}
}
// Scan backward
range.mBegin = 0;
for(i = aOffset ;i > 0; i--)
{
if( c != this->GetClass(aText[i-1]))
{
range.mBegin = i;
break;
}
}
if(kWbClassThaiLetter == c)
{
// need to call Thai word breaker from here
// we should pass the whole Thai segment to the thai word breaker to find a shorter answer
}
return range;
}
int32_t nsSampleWordBreaker::NextWord(
const char16_t* aText, uint32_t aLen, uint32_t aPos)
{
int8_t c1, c2;
uint32_t cur = aPos;
if (cur == aLen)
return NS_WORDBREAKER_NEED_MORE_TEXT;
c1 = this->GetClass(aText[cur]);
for(cur++; cur <aLen; cur++)
{
c2 = this->GetClass(aText[cur]);
if(c2 != c1)
break;
}
if(kWbClassThaiLetter == c1)
{
// need to call Thai word breaker from here
// we should pass the whole Thai segment to the thai word breaker to find a shorter answer
}
if (cur == aLen)
return NS_WORDBREAKER_NEED_MORE_TEXT;
return cur;
}