Merge pull request #33 from hmlendea/japanese

New transliterator for `Japanese`

TransliterationAPI.UnitTests/Service/Transliterators/JapaneseTransliteratorTests.cs

@@ -0,0 +1,62 @@
+using NUnit.Framework;
+using TransliterationAPI.Service.Transliterators;
+
+namespace TransliterationAPI.UnitTests.Service.Transliterators
+{
+    public class JapaneseTransliteratorTests
+    {
+        private IJapaneseTransliterator transliterator;
+
+        [SetUp]
+        public void SetUp()
+        {
+            this.transliterator = new JapaneseTransliterator();
+        }
+
+        [Test]
+        [TestCase("京都", "Kyōto")]
+        [TestCase("仙台", "Sendai")]
+        [TestCase("北海道", "Hokkaidō")]
+        [TestCase("名古屋", "Nagoya")]
+        [TestCase("和歌山", "Wakayama")]
+        [TestCase("大阪", "Ōsaka")]
+        [TestCase("奈良", "Nara")]
+        [TestCase("宮崎", "Miyazaki")]
+        [TestCase("富士山", "Fujisan")]
+        [TestCase("山口", "Yamaguchi")]
+        [TestCase("山形", "Yamagata")]
+        [TestCase("岐阜", "Gifu")]
+        [TestCase("岡山", "Okayama")]
+        [TestCase("島根", "Shimane")]
+        [TestCase("広島", "Hiroshima")]
+        [TestCase("愛媛", "Ehime")]
+        [TestCase("新潟", "Niigata")]
+        [TestCase("札幌", "Sapporo")]
+        [TestCase("東京", "Tōkyō")]
+        [TestCase("横浜", "Yokohama")]
+        [TestCase("横須賀", "Yokosuka")]
+        [TestCase("沖縄", "Okinawa")]
+        [TestCase("滋賀", "Shiga")]
+        [TestCase("熊本", "Kumamoto")]
+        [TestCase("石川", "Ishikawa")]
+        [TestCase("福井", "Fukui")]
+        [TestCase("福岡", "Fukuoka")]
+        [TestCase("福島", "Fukushima")]
+        [TestCase("群馬", "Gunma")]
+        [TestCase("茨城", "Ibaraki")]
+        [TestCase("金沢", "Kanazawa")]
+        [TestCase("鎌倉", "Kamakura")]
+        [TestCase("長崎", "Nagasaki")]
+        [TestCase("長野", "Nagano")]
+        [TestCase("青森", "Aomori")]
+        [TestCase("静岡", "Shizuoka")]
+        [TestCase("高松", "Takamatsu")]
+        [TestCase("高知", "Kōchi")]
+        [TestCase("鳥取", "Tottori")]
+        [TestCase("鹿児島", "Kagoshima")]
+        public void GivenATextInJapaneseScript_WhenTransliteratingIntoLatin_ThenTheCorrectTextIsReturned(
+            string japaneseText,
+            string expectedTransliteratedText)
+            => Assert.That(transliterator.Transliterate(japaneseText), Is.EqualTo(expectedTransliteratedText));
+    }
+}

TransliterationAPI/Service/Transliterators/IJapaneseTransliterator.cs

@@ -0,0 +1,7 @@
+namespace TransliterationAPI.Service.Transliterators
+{
+    public interface IJapaneseTransliterator
+    {
+        string Transliterate(string text);
+    }
+}

TransliterationAPI/Service/Transliterators/JapaneseTransliterator.cs

@@ -0,0 +1,168 @@
+using System.Collections.Generic;
+using System.Text.RegularExpressions;
+
+using NuciExtensions;
+
+namespace TransliterationAPI.Service.Transliterators
+{
+    public class JapaneseTransliterator : IJapaneseTransliterator
+    {
+        Dictionary<char, string> transliterationMap;
+
+        public JapaneseTransliterator()
+        {
+            transliterationMap = new Dictionary<char, string>()
+            {
+                // Basic Hiragana
+                {'あ', "a"}, {'い', "i"}, {'う', "u"}, {'え', "e"}, {'お', "o"},
+                {'か', "ka"}, {'き', "ki"}, {'く', "ku"}, {'け', "ke"}, {'こ', "ko"},
+                {'さ', "sa"}, {'し', "shi"}, {'す', "su"}, {'せ', "se"}, {'そ', "so"},
+                {'た', "ta"}, {'ち', "chi"}, {'つ', "tsu"}, {'て', "te"}, {'と', "to"},
+                {'な', "na"}, {'に', "ni"}, {'ぬ', "nu"}, {'ね', "ne"}, {'の', "no"},
+                {'は', "ha"}, {'ひ', "hi"}, {'ふ', "fu"}, {'へ', "he"}, {'ほ', "ho"},
+                {'ま', "ma"}, {'み', "mi"}, {'む', "mu"}, {'め', "me"}, {'も', "mo"},
+                {'や', "ya"}, {'ゆ', "yu"}, {'よ', "yo"},
+                {'ら', "ra"}, {'り', "ri"}, {'る', "ru"}, {'れ', "re"}, {'ろ', "ro"},
+                {'わ', "wa"}, {'を', "wo"}, {'ん', "n"},
+
+                // Basic Katakana
+                {'ア', "a"}, {'イ', "i"}, {'ウ', "u"}, {'エ', "e"}, {'オ', "o"},
+                {'カ', "ka"}, {'キ', "ki"}, {'ク', "ku"}, {'ケ', "ke"}, {'コ', "ko"},
+                {'サ', "sa"}, {'シ', "shi"}, {'ス', "su"}, {'セ', "se"}, {'ソ', "so"},
+                {'タ', "ta"}, {'チ', "chi"}, {'ツ', "tsu"}, {'テ', "te"}, {'ト', "to"},
+                {'ナ', "na"}, {'ニ', "ni"}, {'ヌ', "nu"}, {'ネ', "ne"}, {'ノ', "no"},
+                {'ハ', "ha"}, {'ヒ', "hi"}, {'フ', "fu"}, {'ヘ', "he"}, {'ホ', "ho"},
+                {'マ', "ma"}, {'ミ', "mi"}, {'ム', "mu"}, {'メ', "me"}, {'モ', "mo"},
+                {'ヤ', "ya"}, {'ユ', "yu"}, {'ヨ', "yo"},
+                {'ラ', "ra"}, {'リ', "ri"}, {'ル', "ru"}, {'レ', "re"}, {'ロ', "ro"},
+                {'ワ', "wa"}, {'ヲ', "wo"}, {'ン', "n"},
+                {'ィ', "i"}, {'デ', "de"}, {'プ', "pu"}, {'グ', "gu"},
+                {'・', " "},
+
+                // Special Characters
+                {'ゃ', "ya"}, {'ゅ', "yu"}, {'ょ', "yo"}, // small ya, yu, yo
+                {'ャ', "ya"}, {'ュ', "yu"}, {'ョ', "yo"}, // small YA, YU, YO
+                {'っ', "tsu"}, // small tsu
+                {'ッ', "tsu"},  // small TSU
+
+                // Kanji for toponyms
+                { '東', "tō" },     // East, as in 東京 (Tōkyō)
+                { '京', "kyō" },    // Capital, as in 京都 (Kyōto)
+                { '大', "ō" },      // Big, as in 大阪 (Ōsaka)
+                { '阪', "saka" },   // Hill, slope, as in 大阪 (Ōsaka)
+                { '北', "hoku" },   // North, as in 北海道 (Hokkaidō)
+                { '海', "kai" },    // Sea, as in 北海道 (Hokkaidō)
+                { '道', "dō" },     // Road, path, as in 北海道 (Hokkaidō)
+                { '名', "na" },     // Name, as in 名古屋 (Nagoya)
+                { '古', "ko" },     // Old, as in 名古屋 (Nagoya)
+                { '屋', "ya" },     // Shop, house, as in 名古屋 (Nagoya)
+                { '神', "kami" },   // God, as in 神戸 (Kōbe)
+                { '戸', "ko" },     // Door, gate, as in 神戸 (Kōbe)
+                { '横', "yoko" },   // Horizontal, as in 横浜 (Yokohama)
+                { '浜', "hama" },   // Beach, as in 横浜 (Yokohama)
+                { '仙', "sen" },    // Hermit, wizard, as in 仙台 (Sendai)
+                { '台', "dai" },    // Stand, support, as in 仙台 (Sendai)
+
+                { '井', "i" },
+                { '倉', "kura" },
+                { '児', "go" },
+                { '取', "tori" },
+                { '口', "guchi" },
+                { '和', "wa" },
+                { '城', "baraki" },
+                { '士', "ji" },
+                { '奈', "na" },
+                { '媛', "hime" },
+                { '宮', "miya" },
+                { '富', "fu" },
+                { '山', "yama" },
+                { '岐', "gi" },
+                { '岡', "oka" },
+                { '島', "shima" },
+                { '崎', "saki" },
+                { '川', "kawa" },
+                { '幌', "poro" },
+                { '広', "hiro" },
+                { '形', "gata" },
+                { '愛', "e" },
+                { '新', "nii" },
+                { '本', "moto" },
+                { '札', "satsu" },
+                { '松', "matsu" },
+                { '根', "ne" },
+                { '森', "mori" },
+                { '歌', "ka" },
+                { '殿', "dono" },
+                { '沖', "oki" },
+                { '沢', "zawa" },
+                { '滋', "shi" },
+                { '潟', "gata" },
+                { '熊', "kuma" },
+                { '知', "chi" },
+                { '石', "ishi" },
+                { '福', "fuku" },
+                { '縄', "nawa" },
+                { '群', "gun" },
+                { '良', "ra" },
+                { '茨', "i" },
+                { '賀', "ka" },
+                { '都', "to" },
+                { '野', "no" },
+                { '金', "kana" },
+                { '鎌', "kama" },
+                { '長', "naga" },
+                { '阜', "fu" },
+                { '青', "ao" },
+                { '静', "shizu" },
+                { '須', "su" },
+                { '馬', "ma" },
+                { '高', "taka" },
+                { '鳥', "tori" },
+                { '鹿', "ka" },
+            };
+
+        }
+
+        public string Transliterate(string text)
+        {
+            string transliteratedText = string.Empty;
+
+            foreach (char character in text)
+            {
+                if (transliterationMap.ContainsKey(character))
+                {
+                    transliteratedText += transliterationMap[character];
+                }
+                else
+                {
+                    transliteratedText += character;
+                }
+            }
+
+            transliteratedText = ApplyFixes(transliteratedText);
+
+            return transliteratedText;
+        }
+
+        string ApplyFixes(string text)
+        {
+            string fixedText = text.ToTitleCase();
+
+            fixedText = Regex.Replace(fixedText, "([Tt])orit", "$1ott");
+            fixedText = Regex.Replace(fixedText, "Takac", "Kōc");
+
+            fixedText = Regex.Replace(fixedText, "Oo", "Ō");
+            fixedText = Regex.Replace(fixedText, "oo", "ō");
+
+            fixedText = Regex.Replace(fixedText, "akoy", "agoy");
+            fixedText = Regex.Replace(fixedText, "ika ", "iga ");
+            fixedText = Regex.Replace(fixedText, "ika$", "iga");
+            fixedText = Regex.Replace(fixedText, "iyama", "isan");
+            fixedText = Regex.Replace(fixedText, "kuk", "kk");
+            fixedText = Regex.Replace(fixedText, "tsup", "pp");
+            fixedText = Regex.Replace(fixedText, "yasa", "yaza");
+
+            return fixedText;
+        }
+    }
+}