|
|
//+--------------------------------------------------------------------------
//
// Microsoft Windows
// Copyright (C) Microsoft Corporation, 1995 - 1999.
//
// File: ctplus.c
//
// Contents: Contains character type (orthography) data and routine
// to get at it.
//
// History: 23-May-96 pathal Created.
//
//---------------------------------------------------------------------------
#include "precomp.h"
#include <winnls.h>
//----------------------------------------------------------------------------
// s_abBreakList
//
// This array starts at -1, so that EOF can be found in the array. It
// depends on (EOF == -1) being true. Also, all references to it must be
// of the form (s_abCharTypeList+1)[x]
//
// 000
// EOF
//
// 001-080
// The lower 7F entries from the ASCII Code Page (0000-00ff) are mapped in place
// (ex. UNICODE 0009 (HT) == 009)
// The word characters are: $,0-9,A-Z,_,a-z
// The word separators are: bs,tab,lf,vtab,cr,spc,
// ",#,%,&,',(,),*,+,comma,-,/,
// :,;,<,=,>,@,[,],`
// The phrase seperators are: !,.,?,\,^,{,|,},~
//
// NOTE: Symbols are treated as WS or PS.
//
// 081-0FF
// The lower 7E entries from the Half Width Variant Code Page (FF00-FF7F) are
// mapped to 081-0FF.
//
// 100-1FF
// The lower FF entries from the General Punctuation Code Page (2000-2044) are
// mapped to 100-1ff.
//
// 200-2FF
// The lower FF entries from the CJK Auxiliary Code Page (3000-30FF) are mapped
// to 200-2ff.
//
// pathal - 5/20/96
// Special default character processing for selection
// The following is a list of white space characters that T-Hammer will not right select on:
// 0x0009 (tab), 0x0020 (ansi space), 0x2005 (narrow space, 0x3000 (wide space)
// (Note: see AnalyzeHPBs for special end SPB processing of adjacent white space)
// The following is a list of nls characters to be treated as text by T-Hammer:
// (in other words T-Hammer will neither right nor left-select on them):
// 0x001F (non-required hyphen), 0x0027 (single quote), 0x2019 (right quote),
// 0x200C (non-width optional break), 0x200D (non-width no break)
//----------------------------------------------------------------------------
const BYTEs_abCharTypeList[0x301] = { (BYTE) -1, // EOF (-1)
PS,PS,PS,PS,PS,PS,PS,PS, WS,WS,WS,WS,PS,WS,PS,PS, // 000 - 015
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,CH,PS, // 016 - 031
WS,PS,WS,WS,CH,CH,WS,PS, // sp ! " # $ % & '
WS,WS,WS,WS,WS,WS,PS,WS, // ( ) * + , - . /
CH,CH,CH,CH,CH,CH,CH,CH, // 0 1 2 3 4 5 6 7
CH,CH,WS,WS,WS,WS,WS,PS, // 8 9 : ; < = > ?
WS,CH,CH,CH,CH,CH,CH,CH, // @ A B C D E F G
CH,CH,CH,CH,CH,CH,CH,CH, // H I J K M L N O
CH,CH,CH,CH,CH,CH,CH,CH, // P Q R S T U V Y
CH,CH,CH,WS,PS,WS,PS,CH, // X Y Z [ \ ] ^ _
WS,CH,CH,CH,CH,CH,CH,CH, // ` a b c d e f g
CH,CH,CH,CH,CH,CH,CH,CH, // h i j k m l n o
CH,CH,CH,CH,CH,CH,CH,CH, // p q r s t u v y
CH,CH,CH,PS,PS,PS,CH,PS, // x y z { | } ~ del
WS,PS,WS,WS,CH,CH,WS,WS, // FF00-FF07 (sp ! " # $ % & ')
WS,WS,WS,WS,WS,WS,PS,WS, // ( ) * + , - . /
VC,VC,VC,VC,VC,VC,VC,VC, // 0 1 2 3 4 5 6 7
VC,VC,WS,WS,WS,WS,WS,PS, // 8 9 : ; < = > ?
WS,VC,VC,VC,VC,VC,VC,VC, // @ A B C D E F G
VC,VC,VC,VC,VC,VC,VC,VC, // H I J K M L N O
VC,VC,VC,VC,VC,VC,VC,VC, // P Q R S T U V Y
VC,VC,VC,WS,VC,WS,PS,VC, // X Y Z [ \ ] ^ _
WS,VC,VC,VC,VC,VC,VC,VC, // ` a b c d e f g
VC,VC,VC,VC,VC,VC,VC,VC, // h i j k m l n o
VC,VC,VC,VC,VC,VC,VC,VC, // p q r s t u v y
VC,VC,VC,PS,PS,PS,VC,PS, // x y z { | } ~ del
VC,PS,WS,WS,WS,WS,VC,VC, // FF60-FF67
VC,VC,VC,VC,VC,VC,VC,VC, // FF68-FF6F
VC,VC,VC,VC,VC,VC,VC,VC, // FF70-FF77
VC,VC,VC,VC,VC,VC,VC,VC, // FF70-FF7E
WS,WS,WS,WS,WS,WS,WS,WS, // 2000-2007
WS,WS,WS,WS,CH,CH,WS,WS, // 2008-200F
WS,CH,WS,WS,WS,KC,PS,WS, // 2010-2017
WS,CH,WS,WS,WS,WS,WS,WS, // 2018-201F
WS,WS,PS,PS,PS,PS,PS,CH, // 2020-2027
PS,PS,CH,CH,CH,CH,CH,PS, // 2028-202F
WS,WS,WS,WS,WS,WS,WS,WS, // 2030-2037
WS,WS,WS,WS,WS,WS,WS,WS, // 2038-203F
WS,WS,WS,PS,WS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2040-204F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2050-205F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2060-206F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2070-207F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2080-208F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 2090-209F
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20A0-20AF
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20B0-20BF
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20C0-20CF
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20D0-20DF
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20E0-20EF
PS,PS,PS,PS,PS,PS,PS,PS, PS,PS,PS,PS,PS,PS,PS,PS, // 20F0-20FF
WS,WS,PS,HC,HC,IC,IC,HC, // 3000-3007
WS,WS,WS,WS,WS,WS,WS,WS, // 3008-300F
WS,WS,WS,WS,WS,WS,WS,WS, // 3010-3017
WS,WS,WS,WS,WS,WS,WS,WS, // 3018-301F
HC,HC,HC,HC,HC,HC,HC,HC, // 3020-3027
HC,HC,HC,HC,HC,HC,HC,HC, // 3028-302F
WS,HC,IC,HC,IC,HC,HC,HC, // 3030-3037
PS,PS,PS,PS,PS,PS,PS,WS, // 3038-303F
WS,HC,HC,HC,HC,HC,HC,HC, HC,HC,HC,HC,HC,HC,HC,HC, // 3040-304F
HC,HC,HC,HC,HC,HC,HC,HC, HC,HC,HC,HC,HC,HC,HC,HC, // 3050-305F
HC,HC,HC,HC,HC,HC,HC,HC, HC,HC,HC,HC,HC,HC,HC,HC, // 3060-306F
HC,HC,HC,HC,HC,HC,HC,HC, HC,HC,HC,HC,HC,HC,HC,HC, // 3070-307F
HC,HC,HC,HC,HC,HC,HC,HC, HC,HC,HC,HC,HC,HC,HC,HC, // 3080-308F
HC,HC,HC,HC,HC,PS,PS,PS, // 3090-3097
PS,HC,HC,WS,WS,HC,HC,PS, // 3098-309F
WS,KC,KC,KC,KC,KC,KC,KC, KC,KC,KC,KC,KC,KC,KC,KC, // 30A0-30AF
KC,KC,KC,KC,KC,KC,KC,KC, KC,KC,KC,KC,KC,KC,KC,KC, // 30B0-30BF
KC,KC,KC,KC,KC,KC,KC,KC, KC,KC,KC,KC,KC,KC,KC,KC, // 30C0-30CF
KC,KC,KC,KC,KC,KC,KC,KC, KC,KC,KC,KC,KC,KC,KC,KC, // 30D0-30DF
KC,KC,KC,KC,KC,KC,KC,KC, KC,KC,KC,KC,KC,KC,KC,KC, // 30E0-30EF
KC,KC,KC,KC,KC,KC,IC,PS, // 30F0-30F7
PS,PS,PS,WS,KC,KC,KC,PS, // 30F8-30FF
};
//
// Type C1 bits are:
//
// C1_UPPER 0x0001 // upper case
// C1_LOWER 0x0002 // lower case
// C1_DIGIT 0x0004 // decimal digits 1
// C1_SPACE 0x0008 // spacing characters 2
// C1_PUNCT 0x0010 // punctuation characters 4
// C1_CNTRL 0x0020 // control characters 8
// C1_BLANK 0x0040 // blank characters 10
// C1_XDIGIT 0x0080 // other digits 20
// C1_ALPHA 0x0100 // any linguistic character 40
//
// But since I don't care about C1_UPPER and C1_LOWER I can right-shift
// the output of GetStringTypeEx and keep a 128 Byte lookup table.
//
// The precedence rules are: (Alpha, XDigit, Digit) --> CH
// (Punct) --> PS
// (Space, Blank, Control) --> WS
//
const BYTEs_abCTypeList[128] = { WS, CH, WS, CH, PS, CH, WS, CH, // 00 - 07
WS, CH, WS, CH, PS, CH, WS, CH, // 08 - 0F
WS, CH, WS, CH, PS, CH, WS, CH, // 10 - 17
WS, CH, WS, CH, PS, CH, WS, CH, // 18 - 1F
CH, CH, CH, CH, CH, CH, CH, CH, // 20 - 27
CH, CH, CH, CH, CH, CH, CH, CH, // 20 - 27
CH, CH, CH, CH, CH, CH, CH, CH, // 30 - 37
CH, CH, CH, CH, CH, CH, CH, CH, // 30 - 37
CH, CH, CH, CH, CH, CH, CH, CH, // 40 - 47
CH, CH, CH, CH, CH, CH, CH, CH, // 48 - 4F
CH, CH, CH, CH, CH, CH, CH, CH, // 50 - 57
CH, CH, CH, CH, CH, CH, CH, CH, // 58 - 5F
CH, CH, CH, CH, CH, CH, CH, CH, // 60 - 67
CH, CH, CH, CH, CH, CH, CH, CH, // 68 - 6F
CH, CH, CH, CH, CH, CH, CH, CH, // 70 - 77
CH, CH, CH, CH, CH, CH, CH, CH, // 78 - 7F
};
//+---------------------------------------------------------------------------
// Member: WBREAKJ::TypeOf
//
// Synopsis: Returns the type of a character
//
// Arguments: [c] -- Unicode Character
//
// Returns: type, one of CH, WS, PS, EOF
//
// History: 08-Apr-94 PatHal Adapted for Japanese WB
//
// Notes: This returns the type of a character, using the static
// array s_abCharTypeList. It adds 1 so that EOF (-1) can be in
// the array, and accessed normally.
//
// This is not done by overloading the [] opeator, because in
// future versions it will not necessarly be a table lookup.
//
// See above (typeof comments) for an explanation of the mapping
//
//----------------------------------------------------------------------------
BYTEGetCharType(WCHAR wc ){ WCHAR wc2;
// Map interesting stuff (0000, 2000, 3000, FF00) to the table range,
// 0x0000 - 0x0300.
//
wc2 = (wc & 0x00FF);
switch (wc & 0xFF00) {
case 0xFF00: // Half-Width Variants
if (wc2 & 0x80) { return(VC); // including Hangul
} wc2 |= 0x0080; break;
case 0xFE00: // Small Variants
if ((wc2 <= 0x006B) && (wc2 != 0x0069)) { return(WS); } // Treat Small $ and arabic symbols as CH
return(CH); // break;
case 0x3000: // CJK Auxiliary
wc2 |= 0x0200; break;
case 0x2000: // General Punctuation
wc2 |= 0x0100; break;
case 0x0000: // Code page 0
// Use System NLS map for code page 0
if (wc2 & 0x80) { WORD wCharType = 0;
GetStringTypeEx( MAKELANGID( LANG_CHINESE, SUBLANG_CHINESE_TRADITIONAL ), CT_CTYPE1, &wc2, 1, &wCharType ); return s_abCTypeList[wCharType >> 2]; } break;
default: //
// Treat the whole CJK Range as Kanji
//
if ((wc >= 0x4E00) && (wc <= 0x9FFF)) { return(IC); }
//
// Treat All Gaiji as Kanji Char, too
//
if ((wc >= 0xE000) && (wc < 0xE758)) { return(IC); }
//
// Treat all CJK symbols as word separators
// NOTE: This means that the stemmer must be smart about searching
// for zipcodes when given one with a preceding zipcode char.
//
if ((wc >= 0x3200) && (wc <= 0x33DD)) { return(WS); }
// If it's not interesting return CH as default;
return(CH); // break;
}
return( (s_abCharTypeList+1)[wc2] );}
|
