출력을 호환자모로 통일하는 옵션 추가

include/kiwi/PatternMatcher.h

@@ -1,4 +1,4 @@
-#pragma once
+#pragma once
 
 #include <vector>
 #include <string>
@@ -24,6 +24,7 @@ namespace kiwi
 		joinAdvSuffix = 1 << 21, /**< 부사파생접미사(XSM)를 분리하지 않고 합쳐서 매칭한다 */
 		splitComplex = 1 << 22, /**< 더 작은 단위로 분할될 수 있는 형태소는 더 분할하여 매칭한다 */
 		zCoda = 1 << 23, /**< 어미 및 조사에 덧붙은 받침이 있는 경우 이를 분리하여 z_coda 태그로 매칭한다 */
+		compatibleJamo = 1 << 24, /**< 출력시 한글 첫가끝 자모를 호환가능한 자모로 변환한다. */
 		joinVSuffix = joinVerbSuffix | joinAdjSuffix,
 		joinAffix = joinNounPrefix | joinNounSuffix | joinVerbSuffix | joinAdjSuffix | joinAdvSuffix,
 		all = url | email | hashtag | mention | serial | emoji | zCoda,

include/kiwi/Utils.h

@@ -58,9 +58,29 @@ namespace kiwi
 		return within(chr, 0xAC00, 0xD7A4);
 	}
 
+	inline bool isHangulOnset(char16_t chr)
+	{
+		return within(chr, 0x1100, 0x1100 + 19);
+	}
+
 	inline bool isHangulCoda(char16_t chr)
 	{
-		return within(chr, 0x11A8, 0x11A7 + 28);
+		return within(chr, 0x11A8, 0x11A8 + 27);
+	}
+
+	inline bool isHangulVowel(char16_t chr)
+	{
+		return within(chr, 0x314F, 0x3164);
+	}
+
+	inline char16_t joinOnsetVowel(size_t onset, size_t vowel)
+	{
+		return 0xAC00 + (char16_t)((onset * 21 + vowel) * 28);
+	}
+
+	inline int extractVowel(char16_t chr)
+	{
+		return ((chr - 0xAC00) / 28) % 21;
 	}
 
 	inline bool isOldHangulOnset(char16_t chr)
@@ -88,6 +108,8 @@ namespace kiwi
 		return within(chr, 0x3131, 0x314E) || within(chr, 0x3165, 0x3186);
 	}
 
+	char16_t toCompatibleHangulConsonant(char16_t chr);
+
 	struct ComparatorIgnoringSpace
 	{
 		static bool less(const KString& a, const KString& b, const kchar_t space = u' ');

src/Kiwi.cpp

@@ -616,6 +616,14 @@ namespace kiwi
 		}
 	}
 
+	inline void toCompatibleJamo(u16string& str)
+	{
+		for (auto& c : str)
+		{
+			c = toCompatibleHangulConsonant(c);
+		}
+	}
+
 	inline void insertPathIntoResults(
 		vector<TokenResult>& ret, 
 		Vector<SpecialState>& spStatesByRet,
@@ -726,6 +734,12 @@ namespace kiwi
 					}
 					joined = joinHangul(s.str.empty() ? *s.morph->kform : s.str);
 				} while (0);
+
+				if (!!(matchOptions & Match::compatibleJamo))
+				{
+					toCompatibleJamo(joined);
+				}
+
 				rarr.emplace_back(joined, s.morph->tag);
 				auto& token = rarr.back();
 				token.morph = within(s.morph, pretokenizedGroup.morphemes) ? nullptr : s.morph;

src/StrUtils.h

@@ -728,21 +728,6 @@ namespace kiwi
 		}
 	}
 
-	inline bool isHangulOnset(char16_t c)
-	{
-		return u'ᄀ' <= c && c <= u'ᄒ';
-	}
-
-	inline bool isHangulVowel(char16_t c)
-	{
-		return u'ㅏ' <= c && c <= u'ㅣ';
-	}
-
-	inline char16_t joinOnsetVowel(size_t onset, size_t vowel)
-	{
-		return u'가' + (char16_t)((onset * 21 + vowel) * 28);
-	}
-
 	inline bool isChineseChr(char32_t c)
 	{
 		return (0x4E00 <= c && c <= 0x9FFF)

src/Utils.cpp

@@ -498,4 +498,16 @@ namespace kiwi
 		return ret;
 	}
 
+	char16_t toCompatibleHangulConsonant(char16_t chr)
+	{
+		if (isHangulOnset(chr))
+		{
+			return u"ㄱㄲㄴㄷㄸㄹㅁㅂㅃㅅㅆㅇㅈㅉㅊㅋㅌㅍㅎ"[chr - 0x1100];
+		}
+		else if (isHangulCoda(chr))
+		{
+			return u"ㄱㄲㄳㄴㄵㄶㄷㄹㄺㄻㄼㄽㄾㄿㅀㅁㅂㅄㅅㅆㅇㅈㅊㅋㅌㅍㅎ"[chr - 0x11A8];
+		}
+		return chr;
+	}
 }

test/test_cpp.cpp

@@ -1228,6 +1228,27 @@ TEST(KiwiCpp, JoinAffix)
 	EXPECT_EQ(res5.first[5].str, u"배송되");
 }
 
+TEST(KiwiCpp, CompatibleJamo)
+{
+	Kiwi& kiwi = reuseKiwiInstance();
+	auto res1 = kiwi.analyze(u"이긴다. 이김. 이길것.", Match::none).first;
+	EXPECT_EQ(res1.size(), 10);
+	EXPECT_EQ(res1[1].str, u"ᆫ다");
+	EXPECT_EQ(res1[4].str, u"ᆷ");
+	EXPECT_EQ(res1[7].str, u"ᆯ");
+
+	auto res2 = kiwi.analyze(u"이긴다. 이김. 이길것.", Match::compatibleJamo).first;
+	EXPECT_EQ(res2.size(), 10);
+	EXPECT_EQ(res2[1].str, u"ㄴ다");
+	EXPECT_EQ(res2[4].str, u"ㅁ");
+	EXPECT_EQ(res2[7].str, u"ㄹ");
+
+	auto res3 = kiwi.analyze(u"ᄀᄁᄂᄃᄄᄅᄆᄇᄈᄉᄊᄋᄌᄍᄎᄏᄐᄑᄒ ᆨᆩᆪᆫᆬᆭᆮᆯᆰᆱᆲᆳᆴᆵᆶᆷᆸᆹᆺᆻᆼᆽᆾᆿᇀᇁᇂ", Match::compatibleJamo).first;
+	EXPECT_EQ(res3.size(), 2);
+	EXPECT_EQ(res3[0].str, u"ㄱㄲㄴㄷㄸㄹㅁㅂㅃㅅㅆㅇㅈㅉㅊㅋㅌㅍㅎ");
+	EXPECT_EQ(res3[1].str, u"ㄱㄲㄳㄴㄵㄶㄷㄹㄺㄻㄼㄽㄾㄿㅀㅁㅂㅄㅅㅆㅇㅈㅊㅋㅌㅍㅎ");
+}
+
 TEST(KiwiCpp, AutoJoiner)
 {
 	Kiwi& kiwi = reuseKiwiInstance();