xbmc/input/InputCodingTableKorean.cpp

   1 /*
   2  *  Copyright (C) 2005-2024 Team Kodi
   3  *  This file is part of Kodi - https://kodi.tv
   4  *
   5  *  SPDX-License-Identifier: GPL-2.0-or-later
   6  *  See LICENSES/README.md for more information.
   7  */
   8
   9 #include "InputCodingTableKorean.h"
  10
  11 #include "utils/CharsetConverter.h"
  12
  13 #include <stdlib.h>
  14
  15 CInputCodingTableKorean::CInputCodingTableKorean() = default;
  16
  17 std::vector<std::wstring> CInputCodingTableKorean::GetResponse(int)
  18 {
  19   return m_words;
  20 }
  21
  22 bool CInputCodingTableKorean::GetWordListPage(const std::string& strCode, bool isFirstPage)
  23 {
  24   return false;
  25 }
  26
  27 void CInputCodingTableKorean::SetTextPrev(const std::string& strTextPrev)
  28 {
  29   m_strTextPrev = strTextPrev;
  30 }
  31
  32 int CInputCodingTableKorean::MergeCode(int choseong, int jungseong, int jongseong)
  33 {
  34   return (unsigned short)0xAC00 + choseong * 21 * 28 + jungseong * 28 + jongseong + 1;
  35 }
  36
  37 // Reference
  38 // https://en.wikipedia.org/wiki/Hangul
  39 // http://www.theyt.net/wiki/%ED%95%9C%EC%98%81%ED%83%80%EB%B3%80%ED%99%98%EA%B8%B0
  40
  41 std::wstring CInputCodingTableKorean::InputToKorean(const std::wstring& input)
  42 {
  43   std::wstring dicEnglish = // L"rRseEfaqQtTdwWczxvgkoiOjpuPhynbml";
  44       {0x72, 0x52, 0x73, 0x65, 0x45, 0x66, 0x61, 0x71, 0x51, 0x74, 0x54,
  45        0x64, 0x77, 0x57, 0x63, 0x7A, 0x78, 0x76, 0x67, 0x6B, 0x6F, 0x69,
  46        0x4F, 0x6A, 0x70, 0x75, 0x50, 0x68, 0x79, 0x6E, 0x62, 0x6D, 0x6C};
  47   std::wstring dicKorean = // L"ㄱㄲㄴㄷㄸㄹㅁㅂㅃㅅㅆㅇㅈㅉㅊㅋㅌㅍㅎㅏㅐㅑㅒㅓㅔㅕㅖㅗㅛㅜㅠㅡㅣ";
  48       {0x3131, 0x3132, 0x3134, 0x3137, 0x3138, 0x3139, 0x3141, 0x3142, 0x3143, 0x3145, 0x3146,
  49        0x3147, 0x3148, 0x3149, 0x314a, 0x314b, 0x314c, 0x314d, 0x314e, 0x314f, 0x3150, 0x3151,
  50        0x3152, 0x3153, 0x3154, 0x3155, 0x3156, 0x3157, 0x315b, 0x315c, 0x3160, 0x3161, 0x3163};
  51   std::wstring dicChoseong = // L"ㄱㄲㄴㄷㄸㄹㅁㅂㅃㅅㅆㅇㅈㅉㅊㅋㅌㅍㅎ";
  52       {0x3131, 0x3132, 0x3134, 0x3137, 0x3138, 0x3139, 0x3141, 0x3142, 0x3143, 0x3145,
  53        0x3146, 0x3147, 0x3148, 0x3149, 0x314a, 0x314b, 0x314c, 0x314d, 0x314e};
  54   std::wstring dicJungseong = // L"ㅏㅐㅑㅒㅓㅔㅕㅖㅗㅘㅙㅚㅛㅜㅝㅞㅟㅠㅡㅢㅣ";
  55       {0x314f, 0x3150, 0x3151, 0x3152, 0x3153, 0x3154, 0x3155, 0x3156, 0x3157, 0x3158, 0x3159,
  56        0x315a, 0x315b, 0x315c, 0x315d, 0x315e, 0x315f, 0x3160, 0x3161, 0x3162, 0x3163};
  57   std::wstring dicJongseong = // L"ㄱㄲㄳㄴㄵㄶㄷㄹㄺㄻㄼㄽㄾㄿㅀㅁㅂㅄㅅㅆㅇㅈㅊㅋㅌㅍㅎ";
  58       {0x3131, 0x3132, 0x3133, 0x3134, 0x3135, 0x3136, 0x3137, 0x3139, 0x313a,
  59        0x313b, 0x313c, 0x313d, 0x313e, 0x313f, 0x3140, 0x3141, 0x3142, 0x3144,
  60        0x3145, 0x3146, 0x3147, 0x3148, 0x314a, 0x314b, 0x314c, 0x314d, 0x314e};
  61
  62   std::wstring korean;
  63
  64   if (input.empty())
  65     return korean;
  66
  67   int choseong = -1, jungseong = -1, jongseong = -1;
  68
  69   for (unsigned int i = 0; i < input.size(); i++)
  70   {
  71     wchar_t ch = input.at(i);
  72     int key = dicKorean.find(ch);
  73
  74     // H/W Keyboard input with English will be changed to Korean
  75     // because H/W input in Korean is not supported.
  76     if (key == -1)
  77       key = dicEnglish.find(ch);
  78
  79     if (key == -1) // If not Korean and English
  80     {
  81       // If there is remained Korean, merge code into character
  82       if (choseong != -1) // There is choseong
  83       {
  84         if (jungseong != -1) // choseong+jungseong+(jongseong)
  85           korean += MergeCode(choseong, jungseong, jongseong);
  86         else // Only choseong
  87           korean += dicChoseong.at(choseong);
  88       }
  89       else
  90       {
  91         if (jungseong != -1) // Jungseong
  92           korean += dicJungseong.at(jungseong);
  93
  94         if (jongseong != -1) // Jongseong
  95           korean += dicJongseong.at(jongseong);
  96       }
  97       choseong = -1;
  98       jungseong = -1;
  99       jongseong = -1;
 100       korean += ch;
 101     }
 102     else if (key < 19) // If key is consonant, key could be choseong or jungseong
 103     {
 104       if (jungseong != -1)
 105       {
 106         // Jungseong without choseong cannot have jongseong.
 107         // So inputted key is jungseong character, new character is begun.
 108         if (choseong == -1)
 109         {
 110           korean += dicJungseong.at(jungseong);
 111           jungseong = -1;
 112           choseong = key;
 113         }
 114         else // Jungseong with choseong can have jongseong.
 115         {
 116           // Chongseong can have two consonant. So this is first consonant of chongseong.
 117           if (jongseong == -1)
 118           {
 119             jongseong = dicJongseong.find(dicKorean.at(key));
 120             if (jongseong == -1) // This consonant cannot be jongseong. ex) "ㄸ", "ㅃ", "ㅉ"
 121             {
 122               korean += MergeCode(choseong, jungseong, jongseong);
 123               choseong = dicChoseong.find(dicKorean.at(key));
 124               jungseong = -1;
 125             }
 126           }
 127           else if (jongseong == 0 && key == 9) // "ㄳ"
 128             jongseong = 2;
 129           else if (jongseong == 3 && key == 12) // "ㄵ"
 130             jongseong = 4;
 131           else if (jongseong == 3 && key == 18) // "ㄶ"
 132             jongseong = 5;
 133           else if (jongseong == 7 && key == 0) // "ㄺ"
 134             jongseong = 8;
 135           else if (jongseong == 7 && key == 6) // "ㄻ"
 136             jongseong = 9;
 137           else if (jongseong == 7 && key == 7) // "ㄼ"
 138             jongseong = 10;
 139           else if (jongseong == 7 && key == 9) // "ㄽ"
 140             jongseong = 11;
 141           else if (jongseong == 7 && key == 16) // "ㄾ"
 142             jongseong = 12;
 143           else if (jongseong == 7 && key == 17) // "ㄿ"
 144             jongseong = 13;
 145           else if (jongseong == 7 && key == 18) // "ㅀ"
 146             jongseong = 14;
 147           else if (jongseong == 16 && key == 9) // "ㅄ"
 148             jongseong = 17;
 149           else // Jongseong is completed. So new consonant is choseong.
 150           {
 151             korean += MergeCode(choseong, jungseong, jongseong);
 152             choseong = dicChoseong.find(dicKorean.at(key));
 153             jungseong = -1;
 154             jongseong = -1;
 155           }
 156         }
 157       }
 158       else // If there is no jungseong, new consonant can be choseong or second part of double
 159           // consonant.
 160       {
 161         // New consonant is choseong. Also it could be first part of double consonant.
 162         if (choseong == -1)
 163         {
 164           // If choseong is already completed, new consonant is another choseong.
 165           // So previous character has only jongseong.
 166           if (jongseong != -1)
 167           {
 168             korean += dicJongseong.at(jongseong);
 169             jongseong = -1;
 170           }
 171           choseong = dicChoseong.find(dicKorean.at(key));
 172         }
 173         // Find double consonant of chongseong
 174         else if (choseong == 0 && key == 9) // "ㄳ"
 175         {
 176           choseong = -1;
 177           jongseong = 2;
 178         }
 179         else if (choseong == 2 && key == 12) // "ㄵ"
 180         {
 181           choseong = -1;
 182           jongseong = 4;
 183         }
 184         else if (choseong == 2 && key == 18) // "ㄶ"
 185         {
 186           choseong = -1;
 187           jongseong = 5;
 188         }
 189         else if (choseong == 5 && key == 0) // "ㄺ"
 190         {
 191           choseong = -1;
 192           jongseong = 8;
 193         }
 194         else if (choseong == 5 && key == 6) // "ㄻ"
 195         {
 196           choseong = -1;
 197           jongseong = 9;
 198         }
 199         else if (choseong == 5 && key == 7) // "ㄼ"
 200         {
 201           choseong = -1;
 202           jongseong = 10;
 203         }
 204         else if (choseong == 5 && key == 9) // "ㄽ"
 205         {
 206           choseong = -1;
 207           jongseong = 11;
 208         }
 209         else if (choseong == 5 && key == 16) // "ㄾ"
 210         {
 211           choseong = -1;
 212           jongseong = 12;
 213         }
 214         else if (choseong == 5 && key == 17) // "ㄿ"
 215         {
 216           choseong = -1;
 217           jongseong = 13;
 218         }
 219         else if (choseong == 5 && key == 18) // "ㅀ"
 220         {
 221           choseong = -1;
 222           jongseong = 14;
 223         }
 224         else if (choseong == 7 && key == 9) // "ㅄ"
 225         {
 226           choseong = -1;
 227           jongseong = 17;
 228         }
 229         else // In this case, previous character has only choseong. And new consonant is choseong.
 230         {
 231           korean += dicChoseong.at(choseong);
 232           choseong = dicChoseong.find(dicKorean.at(key));
 233         }
 234       }
 235     }
 236     else // If key is vowel, key is jungseong.
 237     {
 238       // If previous character has jongseong and this key is jungseong,
 239       // actually latest vowel is not jongseong. It's choseong of new character.
 240       if (jongseong != -1)
 241       {
 242         // If jongseong of previous character is double consonant, we will separate it to two vowel
 243         // again. First part of double consonant is jongseong of previous character. Second part of
 244         // double consonant is choseong of current character.
 245         int newCho;
 246         if (jongseong == 2) // "ㄱ, ㅅ"
 247         {
 248           jongseong = 0;
 249           newCho = 9;
 250         }
 251         else if (jongseong == 4) // "ㄴ, ㅈ"
 252         {
 253           jongseong = 3;
 254           newCho = 12;
 255         }
 256         else if (jongseong == 5) // "ㄴ, ㅎ"
 257         {
 258           jongseong = 3;
 259           newCho = 18;
 260         }
 261         else if (jongseong == 8) // "ㄹ, ㄱ"
 262         {
 263           jongseong = 7;
 264           newCho = 0;
 265         }
 266         else if (jongseong == 9) // "ㄹ, ㅁ"
 267         {
 268           jongseong = 7;
 269           newCho = 6;
 270         }
 271         else if (jongseong == 10) // "ㄹ, ㅂ"
 272         {
 273           jongseong = 7;
 274           newCho = 7;
 275         }
 276         else if (jongseong == 11) // "ㄹ, ㅅ"
 277         {
 278           jongseong = 7;
 279           newCho = 9;
 280         }
 281         else if (jongseong == 12) // "ㄹ, ㅌ"
 282         {
 283           jongseong = 7;
 284           newCho = 16;
 285         }
 286         else if (jongseong == 13) // "ㄹ, ㅍ"
 287         {
 288           jongseong = 7;
 289           newCho = 17;
 290         }
 291         else if (jongseong == 14) // "ㄹ, ㅎ"
 292         {
 293           jongseong = 7;
 294           newCho = 18;
 295         }
 296         else if (jongseong == 17) // "ㅂ, ㅅ"
 297         {
 298           jongseong = 16;
 299           newCho = 9;
 300         }
 301         else
 302         {
 303           // If jongseong is single consonant, previous character has no chongseong.
 304           // It's choseong of current character.
 305           newCho = dicChoseong.find(dicJongseong.at(jongseong));
 306           jongseong = -1;
 307         }
 308         if (choseong != -1) // If previous character has choseong and jungseong.
 309           korean += MergeCode(choseong, jungseong, jongseong);
 310         else // If previous character has Jongseong only.
 311           korean += dicJongseong.at(jongseong);
 312
 313         choseong = newCho;
 314         jungseong = -1;
 315         jongseong = -1;
 316       }
 317       if (jungseong == -1) // If this key is first vowel, it's first part of jungseong.
 318       {
 319         jungseong = dicJungseong.find(dicKorean.at(key));
 320       }
 321       // If there is jungseong already, jungseong is double vowel.
 322       else if (jungseong == 8 && key == 19) // "ㅘ"
 323         jungseong = 9;
 324       else if (jungseong == 8 && key == 20) // "ㅙ"
 325         jungseong = 10;
 326       else if (jungseong == 8 && key == 32) // "ㅚ"
 327         jungseong = 11;
 328       else if (jungseong == 13 && key == 23) // "ㅝ"
 329         jungseong = 14;
 330       else if (jungseong == 13 && key == 24) // "ㅞ"
 331         jungseong = 15;
 332       else if (jungseong == 13 && key == 32) // "ㅟ"
 333         jungseong = 16;
 334       else if (jungseong == 18 && key == 32) // "ㅢ"
 335         jungseong = 19;
 336       else // If two vowel cannot be double vowel.
 337       {
 338         // Previous character is completed.
 339         // Current character is begin with jungseong.
 340         if (choseong != -1)
 341         {
 342           korean += MergeCode(choseong, jungseong, jongseong);
 343           choseong = -1;
 344         }
 345         else // Previous character has jungseon only.
 346           korean += dicJungseong.at(jungseong);
 347         jungseong = -1;
 348         korean += dicKorean.at(key);
 349       }
 350     }
 351   }
 352
 353   // Process last character
 354   if (choseong != -1)
 355   {
 356     if (jungseong != -1) // Current character has choseong and jungseong.
 357       korean += MergeCode(choseong, jungseong, jongseong);
 358     else // Current character has choseong only.
 359       korean += dicChoseong.at(choseong);
 360   }
 361   else
 362   {
 363     if (jungseong != -1) // Current character has jungseong only
 364       korean += dicJungseong.at(jungseong);
 365     else if (jongseong != -1) // Current character has jongseong only
 366       korean += dicJongseong.at(jongseong);
 367   }
 368
 369   return korean;
 370 }
 371
 372 std::string CInputCodingTableKorean::ConvertString(const std::string& strCode)
 373 {
 374   std::wstring input;
 375   std::string result;
 376   g_charsetConverter.utf8ToW(strCode, input);
 377   InputToKorean(input);
 378   g_charsetConverter.wToUTF8(InputToKorean(input), result);
 379   return m_strTextPrev + result;
 380 }