화장실 검색어 매칭 단순화 (#410)

toilet simple search

Author: sanggggg

app-server/subprojects/bounded_context/external_accessibility/application/src/main/kotlin/club/staircrusher/external_accessibility/application/port/in/ExternalAccessibilitySearchService.kt

@@ -31,7 +31,7 @@ class ExternalAccessibilitySearchService(
             }
             .filter {
                 searchText ?: return@filter true
-                searchText.isSimilarWith(it.name)
+                it.name.isSimilarWith(pattern = searchText)
             }
     }
 }

app-server/subprojects/cross_cutting_concern/stdlib/src/main/kotlin/club/staircrusher/stdlib/util/string/StringUtil.kt

@@ -1,120 +1,25 @@
 package club.staircrusher.stdlib.util.string
 
-fun String.emptyToNull() = this.ifBlank { null }
+import java.util.*
 
-// https://en.wikipedia.org/wiki/Levenshtein_distance#:~:text=The%20Levenshtein%20distance%20between%20two,defined%20the%20metric%20in%201965.
-// string similarity by levenshtein distance considering korean
-fun String.isSimilarWith(other: String, maxThreshold: Int = 3): Boolean {
-    val similarity = jamoLevenshtein(this, other)
-    return similarity <= maxThreshold
-}
+fun String.emptyToNull() = this.ifBlank { null }
 
-private fun Char.decomposeHangul(): List<Char>? {
-    val result = mutableListOf<Char>()
-    val choseongs =
-        listOf('ㄱ', 'ㄲ', 'ㄴ', 'ㄷ', 'ㄸ', 'ㄹ', 'ㅁ', 'ㅂ', 'ㅃ', 'ㅅ', 'ㅆ', 'ㅇ', 'ㅈ', 'ㅉ', 'ㅊ', 'ㅋ', 'ㅌ', 'ㅍ', 'ㅎ')
-    val joongseongs =
-        listOf('ㅏ', 'ㅐ', 'ㅑ', 'ㅒ', 'ㅓ', 'ㅔ', 'ㅕ', 'ㅖ', 'ㅗ', 'ㅘ', 'ㅙ', 'ㅚ', 'ㅛ', 'ㅜ', 'ㅝ', 'ㅞ', 'ㅟ', 'ㅠ', 'ㅡ', 'ㅢ', 'ㅣ')
-    val jongseongs = listOf(
-        null,
-        'ㄱ',
-        'ㄲ',
-        'ㄳ',
-        'ㄴ',
-        'ㄵ',
-        'ㄶ',
-        'ㄷ',
-        'ㄹ',
-        'ㄺ',
-        'ㄻ',
-        'ㄼ',
-        'ㄽ',
-        'ㄾ',
-        'ㄿ',
-        'ㅀ',
-        'ㅁ',
-        'ㅂ',
-        'ㅄ',
-        'ㅅ',
-        'ㅆ',
-        'ㅇ',
-        'ㅈ',
-        'ㅊ',
-        'ㅋ',
-        'ㅌ',
-        'ㅍ',
-        'ㅎ'
+fun String.isSimilarWith(pattern: String): Boolean {
+    return simpleMatch(
+        this.lowercase(Locale.US),
+        pattern.lowercase(Locale.US).filter { it.isWhitespace().not() }
     )
-    val char = this
-    val codePoint = char.code
-
-    if (codePoint in 44032..55203) {
-        val baseCode = codePoint - 44032
-        val choseongIndex = baseCode / 21 / 28
-        val joongseongIndex = baseCode / 28 % 21
-        val jongseongIndex = baseCode % 28
-
-        result.addAll(
-            listOfNotNull(choseongs[choseongIndex], joongseongs[joongseongIndex], jongseongs[jongseongIndex]),
-        )
-    } else {
-        return null
-    }
-    return result
 }
 
-
-private fun levenshtein(s1: String, s2: String, cost: Map<Pair<Char, Char>, Int> = emptyMap()): Int {
-    if (s1.length < s2.length) {
-        return levenshtein(s2, s1, cost)
-    }
-
-    if (s2.isEmpty()) {
-        return s1.length
-    }
-
-    val previousRow = IntArray(s2.length + 1) { it }
-    for (i in s1.indices) {
-        val currentRow = IntArray(s2.length + 1)
-        currentRow[0] = i + 1
-        for (j in s2.indices) {
-            val insertion = previousRow[j + 1] + 1
-            val deletion = currentRow[j] + 1
-            val substitution = previousRow[j] + if (s1[i] == s2[j]) 0 else cost.getOrDefault(s1[i] to s2[j], 1)
-            currentRow[j + 1] = minOf(insertion, deletion, substitution)
+private fun simpleMatch(text: String, pattern: String): Boolean {
+    var patternIndex = 0
+    for (char in text) {
+        if (patternIndex < pattern.length && pattern[patternIndex] == char) {
+            patternIndex++
         }
-        previousRow.indices.forEach { previousRow[it] = currentRow[it] } // Optimized copy
-    }
-    return previousRow.last()
-}
-
-private fun jamoLevenshtein(s1: String, s2: String): Int {
-    if (s1.length < s2.length) {
-        return jamoLevenshtein(s2, s1)
-    }
-
-    if (s2.isEmpty()) {
-        return s1.length
-    }
-
-    val previousRow = IntArray(s2.length + 1) { it }
-    for (i in s1.indices) {
-        val currentRow = IntArray(s2.length + 1)
-        currentRow[0] = i + 1
-        for (j in s2.indices) {
-            val insertion = previousRow[j + 1] + 1
-            val deletion = currentRow[j] + 1
-            val substitution = previousRow[j] + getJamoCost(s1[i], s2[j])
-            currentRow[j + 1] = minOf(insertion, deletion, substitution)
+        if (patternIndex == pattern.length) {
+            return true
         }
-        previousRow.indices.forEach { previousRow[it] = currentRow[it] } // Optimized copy
     }
-    return previousRow.last()
-}
-
-private fun getJamoCost(c1: Char, c2: Char): Int {
-    if (c1 == c2) return 0
-    val jamo1 = c1.decomposeHangul()
-    val jamo2 = c2.decomposeHangul()
-    return if (jamo1 != null && jamo2 != null) levenshtein(jamo1.joinToString(""), jamo2.joinToString("")) / 3 else 1
+    return false
 }

app-server/subprojects/cross_cutting_concern/stdlib/src/unitTest/kotlin/club/staircrusher/stdlib/geography/StringUtilTest.kt

@@ -10,11 +10,11 @@ class StringUtilTest {
         assert("농협하나로마트".isSimilarWith("농하나로마트"))
         assert("농협하나로마트".isSimilarWith("하나로마트"))
         assert("농협하나로마트".isSimilarWith("농협마트"))
+        assert("농협하나로마트".isSimilarWith("농협"))
         assert("NonghyupMart".isSimilarWith("NonghyupMart"))
-        assert("농협하나로마트".isSimilarWith("농협허너루미틋"))
+        assert("NonghyupMart".isSimilarWith("nonghyup"))
 
         assert(!"농협하나로마트".isSimilarWith("아무런"))
         assert(!"농협하나로마트".isSimilarWith("asdf"))
-        assert(!"농협하나로마트".isSimilarWith("농협"))
     }
 }