WIP

flake.lock

@@ -3,19 +3,19 @@
     "jmdict-src": {
       "flake": false,
       "locked": {
-        "narHash": "sha256-+hXdq4aUDuxpCoLJ4zYuKg6zQLHas+/M6+sEDA84RgQ=",
+        "narHash": "sha256-J7GywcE/5LD6YFhdIXxQ2kvAamYkMpannRQyj5yU/nA=",
         "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict.gz"
+        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz"
       },
       "original": {
         "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict.gz"
+        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz"
       }
     },
     "jmdict-with-examples-src": {
       "flake": false,
       "locked": {
-        "narHash": "sha256-H3a5XoSJUvzTReP5g3jL7JGTIEsGyMrLopV4Yfxewig=",
+        "narHash": "sha256-cQtnDLqSHP1be6rkK2ceHL3HmX4YnfvUQi/af3uBQDc=",
         "type": "file",
         "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e_examp.gz"
       },
@@ -27,7 +27,7 @@
     "kanjidic2-src": {
       "flake": false,
       "locked": {
-        "narHash": "sha256-sQS2knH7D/qqrn1BchAvFwjajp9HXv/5r4jx1OEuJWs=",
+        "narHash": "sha256-ml7ZfAXmjDD1sVKhHMS66ytFItIjXRIppqSaYu8IAws=",
         "type": "file",
         "url": "https://www.edrdg.org/kanjidic/kanjidic2.xml.gz"
       },
@@ -38,11 +38,11 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1745391562,
-        "narHash": "sha256-sPwcCYuiEopaafePqlG826tBhctuJsLx/mhKKM5Fmjo=",
+        "lastModified": 1745526057,
+        "narHash": "sha256-ITSpPDwvLBZBnPRS2bUcHY3gZSwis/uTe255QgMtTLA=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "8a2f738d9d1f1d986b5a4cd2fd2061a7127237d7",
+        "rev": "f771eb401a46846c1aebd20552521b233dd7e18b",
         "type": "github"
       },
       "original": {

flake.nix

@@ -5,7 +5,8 @@
     nixpkgs.url = "nixpkgs/nixos-unstable";
 
     jmdict-src = {
-      url = "http://ftp.edrdg.org/pub/Nihongo/JMdict.gz";
+      # url = "http://ftp.edrdg.org/pub/Nihongo/JMdict.gz";
+      url = "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz";
       flake = false;
     };
 

lib/_data_ingestion/jmdict/objects.dart

@@ -15,9 +15,9 @@ abstract class TableNames {
   static const String senseField = 'JMdict_SenseField';
   static const String senseGlossary = 'JMdict_SenseGlossary';
   static const String senseInfo = 'JMdict_SenseInfo';
-  static const String senseLanguageSource = 'JMdict_SenseLanguageSource';
   static const String senseMisc = 'JMdict_SenseMisc';
   static const String sensePOS = 'JMdict_SensePOS';
+  static const String senseLanguageSource = 'JMdict_SenseLanguageSource';
   static const String senseRestrictedToKanji = 'JMdict_SenseRestrictedToKanji';
   static const String senseRestrictedToReading =
       'JMdict_SenseRestrictedToReading';
@@ -216,7 +216,22 @@ class Sense extends SQLWritable {
   });
 
   @override
-  Map<String, Object?> get sqlValue => {};
+  Map<String, Object?> get sqlValue => {
+        'id': id,
+        'orderNum': orderNum,
+      };
+
+  bool get isEmpty => antonyms.isEmpty &&
+      dialects.isEmpty &&
+      fields.isEmpty &&
+      info.isEmpty &&
+      languageSource.isEmpty &&
+      glossary.isEmpty &&
+      misc.isEmpty &&
+      pos.isEmpty &&
+      restrictedToKanji.isEmpty &&
+      restrictedToReading.isEmpty &&
+      seeAlso.isEmpty;
 }
 
 class Entry extends SQLWritable {

lib/_data_ingestion/jmdict/parser.dart

@@ -196,58 +196,65 @@ List<Entry> parseXML(XmlElement root) {
 
     for (final (orderNum, sense) in entry.findElements('sense').indexed) {
       senseId++;
-      senses.add(
-        Sense(
-          id: senseId,
-          orderNum: orderNum + 1,
-          restrictedToKanji:
-              sense.findElements('stagk').map((e) => e.innerText).toList(),
-          restrictedToReading:
-              sense.findElements('stagr').map((e) => e.innerText).toList(),
-          pos: sense
-              .findElements('pos')
-              .map((e) => e.innerText.substring(1, e.innerText.length - 1))
-              .toList(),
-          misc: sense
-              .findElements('misc')
-              .map((e) => e.innerText.substring(1, e.innerText.length - 1))
-              .toList(),
-          dialects: sense
-              .findElements('dial')
-              .map((e) => e.innerText.substring(1, e.innerText.length - 1))
-              .toList(),
-          info: sense.findElements('s_inf').map((e) => e.innerText).toList(),
-          languageSource: sense
-              .findElements('lsource')
-              .map(
-                (e) => LanguageSource(
-                  language: e.getAttribute('xml:lang') ?? 'eng',
-                  fullyDescribesSense: e.getAttribute('ls_type') == 'part',
-                  constructedFromSmallerWords:
-                      e.getAttribute('ls_wasei') == 'y',
-                ),
-              )
-              .toList(),
-          glossary: sense
-              .findElements('gloss')
-              .map(
-                (e) => Glossary(
-                  language: e.getAttribute('xml:lang') ?? 'eng',
-                  phrase: e.innerText,
-                  type: e.getAttribute('g_type'),
-                ),
-              )
-              .toList(),
-          antonyms: sense
-              .findElements('ant')
-              .map((e) => parseXrefParts(e.innerText))
-              .toList(),
-          seeAlso: sense
-              .findElements('xref')
-              .map((e) => parseXrefParts(e.innerText))
-              .toList(),
-        ),
+      final result = Sense(
+        id: senseId,
+        orderNum: orderNum + 1,
+        restrictedToKanji:
+            sense.findElements('stagk').map((e) => e.innerText).toList(),
+        restrictedToReading:
+            sense.findElements('stagr').map((e) => e.innerText).toList(),
+        pos: sense
+            .findElements('pos')
+            .map((e) => e.innerText.substring(1, e.innerText.length - 1))
+            .toList(),
+        misc: sense
+            .findElements('misc')
+            .map((e) => e.innerText.substring(1, e.innerText.length - 1))
+            .toList(),
+        dialects: sense
+            .findElements('dial')
+            .map((e) => e.innerText.substring(1, e.innerText.length - 1))
+            .toList(),
+        info: sense.findElements('s_inf').map((e) => e.innerText).toList(),
+        languageSource: sense
+            .findElements('lsource')
+            .map(
+              (e) => LanguageSource(
+                language: e.getAttribute('xml:lang') ?? 'eng',
+                fullyDescribesSense: e.getAttribute('ls_type') == 'part',
+                constructedFromSmallerWords: e.getAttribute('ls_wasei') == 'y',
+              ),
+            )
+            .toList(),
+        glossary: sense
+            .findElements('gloss')
+            .map(
+              (e) => Glossary(
+                language: e.getAttribute('xml:lang') ?? 'eng',
+                phrase: e.innerText,
+                type: e.getAttribute('g_type'),
+              ),
+            )
+            .toList(),
+        antonyms: sense
+            .findElements('ant')
+            .map((e) => parseXrefParts(e.innerText))
+            .toList(),
+        seeAlso: sense
+            .findElements('xref')
+            .map((e) => parseXrefParts(e.innerText))
+            .toList(),
       );
+
+      if (result.isEmpty) {
+        print(
+          'WARNING: Sense $senseId for entry $entryId is empty, '
+          'kanji: ${kanjiEls.map((e) => e.reading).join(', ')}, '
+          'reading: ${readingEls.map((e) => e.reading).join(', ')}',
+        );
+      } else {
+        senses.add(result);
+      }
     }
 
     entries.add(
@@ -264,7 +271,7 @@ List<Entry> parseXML(XmlElement root) {
 }
 
 Future<void> insertIntoDB(List<Entry> entries, Database db) async {
-  print('  [JMdict] Batch 1');
+  print('  [JMdict] Batch 1 - Kanji and readings');
   Batch b = db.batch();
   for (final e in entries) {
     b.insert(TableNames.entry, e.sqlValue);
@@ -326,32 +333,26 @@ Future<void> insertIntoDB(List<Entry> entries, Database db) async {
 
     for (final s in e.senses) {
       for (final g in s.glossary) {
-        if (g.language == "eng")
-          b.insert(
-            TableNames.entryByEnglish,
-            {
-              'entryId': e.id,
-              'english': g.phrase,
-            },
-            // Some entries have the same reading twice with difference in katakana and hiragana
-            conflictAlgorithm: ConflictAlgorithm.ignore,
-          );
+        b.insert(
+          TableNames.entryByEnglish,
+          {
+            'entryId': e.id,
+            'english': g.phrase,
+          },
+          // Some entries have the same reading twice with difference in katakana and hiragana
+          conflictAlgorithm: ConflictAlgorithm.ignore,
+        );
       }
     }
   }
 
   await b.commit();
 
-  print('  [JMdict] Batch 2');
+  print('  [JMdict] Batch 2 - Senses');
   b = db.batch();
-
   for (final e in entries) {
     for (final s in e.senses) {
-      b.insert(
-          TableNames.sense,
-          s.sqlValue
-            ..addAll({'id': s.id, 'entryId': e.id, 'orderNum': s.orderNum}));
-
+      b.insert(TableNames.sense, s.sqlValue..addAll({'entryId': e.id}));
       for (final d in s.dialects) {
         b.insert(TableNames.senseDialect, {'senseId': s.id, 'dialect': d});
       }
@@ -392,21 +393,17 @@ Future<void> insertIntoDB(List<Entry> entries, Database db) async {
         );
       }
       for (final g in s.glossary) {
-        if (g.language == 'eng')
-          b.insert(
-            TableNames.senseGlossary,
-            g.sqlValue..addAll({'senseId': s.id}),
-            // There are some duplicate glossary, especially in
-            // the other languages.
-            conflictAlgorithm: ConflictAlgorithm.ignore,
-          );
+        b.insert(
+          TableNames.senseGlossary,
+          g.sqlValue..addAll({'senseId': s.id}),
+        );
       }
     }
   }
 
   await b.commit();
 
-  print('  [JMdict] Building trees');
+  print('  [JMdict] Building xref trees');
   SplayTreeMap<String, Set<Entry>> entriesByKanji = SplayTreeMap();
   for (final entry in entries) {
     for (final kanji in entry.kanji) {
@@ -428,7 +425,7 @@ Future<void> insertIntoDB(List<Entry> entries, Database db) async {
     }
   }
 
-  print('  [JMdict] Batch 3');
+  print('  [JMdict] Batch 3 - Xrefs');
   b = db.batch();
 
   for (final e in entries) {

lib/_data_ingestion/sql_writable.dart

@@ -1,5 +1,11 @@
+/// Interface for objects which are meant to be written to a table in a SQL database.
 abstract class SQLWritable {
   const SQLWritable();
 
+  /// Returns a map of the object's properties and their values.
+  /// 
+  /// Note that there might be properties in the object which is meant to be
+  /// inserted into a different table. These properties will/should be excluded
+  /// from this map.
   Map<String, Object?> get sqlValue;
 }

lib/models/jmdict/jmdict_dialect.dart

@@ -0,0 +1,43 @@
+/// Dialect tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_dial
+enum JMdictDialect {
+  brazilian(id: 'bra', description: 'Brazilian'),
+  hokkaido(id: 'hob', description: 'Hokkaido-ben'),
+  kansai(id: 'ksb', description: 'Kansai-ben'),
+  kantou(id: 'ktb', description: 'Kantou-ben'),
+  kyoto(id: 'kyb', description: 'Kyoto-ben'),
+  kyushu(id: 'kyu', description: 'Kyuushuu-ben'),
+  nagano(id: 'nab', description: 'Nagano-ben'),
+  osaka(id: 'osb', description: 'Osaka-ben'),
+  ryukyu(id: 'rkb', description: 'Ryuukyuu-ben'),
+  tokyo(id: 'std', description: 'Tokyo-ben (std)'),
+  tohoku(id: 'thb', description: 'Touhoku-ben'),
+  tosa(id: 'tsb', description: 'Tosa-ben'),
+  tsugaru(id: 'tsug', description: 'Tsugaru-ben');
+
+  final String id;
+  final String description;
+
+  const JMdictDialect({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictDialect fromId(String id) =>
+      JMdictDialect.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictDialect fromJson(Map<String, Object?> json) =>
+      JMdictDialect.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/jmdict_field.dart

@@ -0,0 +1,126 @@
+/// Domain / Field of expertise tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_fld
+enum JMdictField {
+  agriculture(id: 'agric', description: 'agriculture'),
+  anatomy(id: 'anat', description: 'anatomy'),
+  archeology(id: 'archeol', description: 'archeology'),
+  architecture(id: 'archit', description: 'architecture'),
+  art(id: 'art', description: 'art, aesthetics'),
+  astronomy(id: 'astron', description: 'astronomy'),
+  audiovisual(id: 'audvid', description: 'audiovisual'),
+  aviation(id: 'aviat', description: 'aviation'),
+  baseball(id: 'baseb', description: 'baseball'),
+  biochemistry(id: 'biochem', description: 'biochemistry'),
+  biology(id: 'biol', description: 'biology'),
+  botany(id: 'bot', description: 'botany'),
+  boxing(id: 'boxing', description: 'boxing'),
+  buddhism(id: 'Buddh', description: 'Buddhism'),
+  business(id: 'bus', description: 'business'),
+  cardGames(id: 'cards', description: 'card games'),
+  chemistry(id: 'chem', description: 'chemistry'),
+  chineseMythology(id: 'chmyth', description: 'Chinese mythology'),
+  christianity(id: 'Christn', description: 'Christianity'),
+  civilEngineering(id: 'civeng', description: 'civil engineering'),
+  clothing(id: 'cloth', description: 'clothing'),
+  computing(id: 'comp', description: 'computing'),
+  crystallography(id: 'cryst', description: 'crystallography'),
+  dentistry(id: 'dent', description: 'dentistry'),
+  ecology(id: 'ecol', description: 'ecology'),
+  economics(id: 'econ', description: 'economics'),
+  electricEngineering(id: 'elec', description: 'electricity, elec. eng.'),
+  electronics(id: 'electr', description: 'electronics'),
+  embryology(id: 'embryo', description: 'embryology'),
+  engineering(id: 'engr', description: 'engineering'),
+  entomology(id: 'ent', description: 'entomology'),
+  figureSkating(id: 'figskt', description: 'figure skating'),
+  film(id: 'film', description: 'film'),
+  finance(id: 'finc', description: 'finance'),
+  fishing(id: 'fish', description: 'fishing'),
+  food(id: 'food', description: 'food, cooking'),
+  gardening(id: 'gardn', description: 'gardening, horticulture'),
+  genetics(id: 'genet', description: 'genetics'),
+  geography(id: 'geogr', description: 'geography'),
+  geology(id: 'geol', description: 'geology'),
+  geometry(id: 'geom', description: 'geometry'),
+  go(id: 'go', description: 'go (game)'),
+  golf(id: 'golf', description: 'golf'),
+  grammar(id: 'gramm', description: 'grammar'),
+  greekMythology(id: 'grmyth', description: 'Greek mythology'),
+  hanafuda(id: 'hanaf', description: 'hanafuda'),
+  horseRacing(id: 'horse', description: 'horse racing'),
+  internet(id: 'internet', description: 'Internet'),
+  japaneseMythology(id: 'jpmyth', description: 'Japanese mythology'),
+  kabuki(id: 'kabuki', description: 'kabuki'),
+  law(id: 'law', description: 'law'),
+  linguistics(id: 'ling', description: 'linguistics'),
+  logic(id: 'logic', description: 'logic'),
+  martialArts(id: 'MA', description: 'martial arts'),
+  mahjong(id: 'mahj', description: 'mahjong'),
+  manga(id: 'manga', description: 'manga'),
+  mathematics(id: 'math', description: 'mathematics'),
+  mechanicalEngineering(id: 'mech', description: 'mechanical engineering'),
+  medicine(id: 'med', description: 'medicine'),
+  meteorology(id: 'met', description: 'meteorology'),
+  military(id: 'mil', description: 'military'),
+  mineralogy(id: 'min', description: 'mineralogy'),
+  mining(id: 'mining', description: 'mining'),
+  motorsport(id: 'motor', description: 'motorsport'),
+  music(id: 'music', description: 'music'),
+  noh(id: 'noh', description: 'noh (theatre)'),
+  ornithology(id: 'ornith', description: 'ornithology'),
+  paleontology(id: 'paleo', description: 'paleontology'),
+  pathology(id: 'pathol', description: 'pathology'),
+  pharmacology(id: 'pharm', description: 'pharmacology'),
+  philosophy(id: 'phil', description: 'philosophy'),
+  photography(id: 'photo', description: 'photography'),
+  physics(id: 'physics', description: 'physics'),
+  physiology(id: 'physiol', description: 'physiology'),
+  politics(id: 'politics', description: 'politics'),
+  printing(id: 'print', description: 'printing (press)'),
+  professionalWrestling(id: 'prowres', description: 'professional wrestling'),
+  psychiatry(id: 'psy', description: 'psychiatry'),
+  psychoanalysis(id: 'psyanal', description: 'psychoanalysis'),
+  psychology(id: 'psych', description: 'psychology'),
+  railway(id: 'rail', description: 'railway'),
+  romanMythology(id: 'rommyth', description: 'Roman mythology'),
+  shinto(id: 'Shinto', description: 'Shinto'),
+  shogi(id: 'shogi', description: 'shogi (game)'),
+  skiing(id: 'ski', description: 'skiing'),
+  sports(id: 'sports', description: 'sports'),
+  statistics(id: 'stat', description: 'statistics'),
+  stockMarket(id: 'stockm', description: 'stock market'),
+  sumo(id: 'sumo', description: 'sumo (wrestling)'),
+  surgery(id: 'surg', description: 'surgery'),
+  telecommunications(id: 'telec', description: 'telecommunications'),
+  trademark(id: 'tradem', description: 'trademark'),
+  television(id: 'tv', description: 'television'),
+  veterinaryTerms(id: 'vet', description: 'veterinary terms'),
+  videoGames(id: 'vidg', description: 'video games'),
+  zoology(id: 'zool', description: 'zoology');
+
+  final String id;
+  final String description;
+
+  const JMdictField({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictField fromId(String id) =>
+      JMdictField.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictField fromJson(Map<String, Object?> json) =>
+      JMdictField.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/jmdict_kanji_info.dart

@@ -0,0 +1,36 @@
+/// Kanji info tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_kinf
+enum JMdictKanjiInfo {
+  ateji(id: 'ateji', description: 'ateji (phonetic) reading'),
+  ik(id: 'ik', description: 'word containing irregular kana usage'),
+  io(id: 'io', description: 'irregular okurigana usage'),
+  ok(id: 'oK', description: 'word containing out-dated kanji or kanji usage'),
+  rk(id: 'rK', description: 'rarely-used kanji form'),
+  sk(id: 'sK', description: 'search-only kanji form');
+
+  final String id;
+  final String description;
+
+  const JMdictKanjiInfo({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictKanjiInfo fromId(String id) =>
+      JMdictKanjiInfo.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictKanjiInfo fromJson(Map<String, Object?> json) =>
+      JMdictKanjiInfo.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/jmdict_misc.dart

@@ -0,0 +1,98 @@
+/// Miscellaneous sense tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_misc
+enum JMdictMisc {
+  abbreviation(id: 'abbr', description: 'abbreviation'),
+  aphorism(id: 'aphorism', description: 'aphorism (pithy saying)'),
+  archaic(id: 'arch', description: 'archaic'),
+  character(id: 'char', description: 'character'),
+  childrensLanguage(id: 'chn', description: 'children\'s language'),
+  colloquial(id: 'col', description: 'colloquial'),
+  company(id: 'company', description: 'company name'),
+  creature(id: 'creat', description: 'creature'),
+  datedTerm(id: 'dated', description: 'dated term'),
+  deity(id: 'dei', description: 'deity'),
+  derogatory(id: 'derog', description: 'derogatory'),
+  document(id: 'doc', description: 'document'),
+  euphemistic(id: 'euph', description: 'euphemistic'),
+  event(id: 'ev', description: 'event'),
+  familiarLanguage(id: 'fam', description: 'familiar language'),
+  female(id: 'fem', description: 'female term, language, or name'),
+  fiction(id: 'fict', description: 'fiction'),
+  formal(id: 'form', description: 'formal or literary term'),
+  givenName(
+    id: 'given',
+    description: 'given name or forename, gender not specified',
+  ),
+  group(id: 'group', description: 'group'),
+  historical(id: 'hist', description: 'historical term'),
+  honorific(
+    id: 'hon',
+    description: 'honorific or respectful (sonkeigo) language',
+  ),
+  humble(id: 'hum', description: 'humble (kenjougo) language'),
+  idiomatic(id: 'id', description: 'idiomatic expression'),
+  jocular(id: 'joc', description: 'jocular, humorous term'),
+  legend(id: 'leg', description: 'legend'),
+  mangaSlang(id: 'm-sl', description: 'manga slang'),
+  male(id: 'male', description: 'male term, language, or name'),
+  mythology(id: 'myth', description: 'mythology'),
+  internetSlang(id: 'net-sl', description: 'Internet slang'),
+  object(id: 'obj', description: 'object'),
+  obsolete(id: 'obs', description: 'obsolete term'),
+  onomatopoeic(id: 'on-mim', description: 'onomatopoeic or mimetic word'),
+  organizationName(id: 'organization', description: 'organization name'),
+  other(id: 'oth', description: 'other'),
+  personName(id: 'person', description: 'full name of a particular person'),
+  placeName(id: 'place', description: 'place name'),
+  poetical(id: 'poet', description: 'poetical term'),
+  polite(id: 'pol', description: 'polite (teineigo) language'),
+  product(id: 'product', description: 'product name'),
+  proverb(id: 'proverb', description: 'proverb'),
+  quotation(id: 'quote', description: 'quotation'),
+  rare(id: 'rare', description: 'rare term'),
+  religion(id: 'relig', description: 'religion'),
+  sensitive(id: 'sens', description: 'sensitive'),
+  service(id: 'serv', description: 'service'),
+  ship(id: 'ship', description: 'ship name'),
+  slang(id: 'sl', description: 'slang'),
+  railwayStation(id: 'station', description: 'railway station'),
+  surname(id: 'surname', description: 'family or surname'),
+  onlyKana(id: 'uk', description: 'word usually written using kana alone'),
+  unclassifiedName(id: 'unclass', description: 'unclassified name'),
+  vulgar(id: 'vulg', description: 'vulgar expression or word'),
+  workOfArt(
+    id: 'work',
+    description: 'work of art, literature, music, etc. name',
+  ),
+  xRated(
+    id: 'X',
+    description: 'rude or X-rated term (not displayed in educational software)',
+  ),
+  yojijukugo(id: 'yoji', description: 'yojijukugo');
+
+  final String id;
+  final String description;
+
+  const JMdictMisc({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictMisc fromId(String id) =>
+      JMdictMisc.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictMisc fromJson(Map<String, Object?> json) =>
+      JMdictMisc.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/jmdict_pos.dart

@@ -0,0 +1,122 @@
+/// Part-of-speech tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_pos
+enum JMdictPOS {
+  adjF(id: 'adj-f', description: 'noun or verb acting prenominally'),
+  adjI(id: 'adj-i', description: 'adjective (keiyoushi)'),
+  adjIx(id: 'adj-ix', description: 'adjective (keiyoushi) - yoi/ii class'),
+  adjKari(id: 'adj-kari', description: '\'kari\' adjective (archaic)'),
+  adjKu(id: 'adj-ku', description: '\'ku\' adjective (archaic)'),
+  adjNa(id: 'adj-na', description: 'adjectival nouns or quasi-adjectives (keiyodoshi)'),
+  adjNari(id: 'adj-nari', description: 'archaic/formal form of na-adjective'),
+  adjNo(id: 'adj-no', description: 'nouns which may take the genitive case particle ''no'''),
+  adjPn(id: 'adj-pn', description: 'pre-noun adjectival (rentaishi)'),
+  adjShiku(id: 'adj-shiku', description: '\'shiku\' adjective (archaic)'),
+  adjT(id: 'adj-t', description: '\'taru\' adjective'),
+  adv(id: 'adv', description: 'adverb (fukushi)'),
+  advTo(id: 'adv-to', description: 'adverb taking the ''to'' particle'),
+  aux(id: 'aux', description: 'auxiliary'),
+  auxAdj(id: 'aux-adj', description: 'auxiliary adjective'),
+  auxV(id: 'aux-v', description: 'auxiliary verb'),
+  conj(id: 'conj', description: 'conjunction'),
+  cop(id: 'cop', description: 'copula'),
+  ctr(id: 'ctr', description: 'counter'),
+  exp(id: 'exp', description: 'expressions (phrases, clauses, etc.)'),
+  int(id: 'int', description: 'interjection (kandoushi)'),
+  n(id: 'n', description: 'noun (common) (futsuumeishi)'),
+  nAdv(id: 'n-adv', description: 'adverbial noun (fukushitekimeishi)'),
+  nPr(id: 'n-pr', description: 'proper noun'),
+  nPref(id: 'n-pref', description: 'noun, used as a prefix'),
+  nSuf(id: 'n-suf', description: 'noun, used as a suffix'),
+  nT(id: 'n-t', description: 'noun (temporal) (jisoumeishi)'),
+  num(id: 'num', description: 'numeric'),
+  pn(id: 'pn', description: 'pronoun'),
+  pref(id: 'pref', description: 'prefix'),
+  prt(id: 'prt', description: 'particle'),
+  suf(id: 'suf', description: 'suffix'),
+  unc(id: 'unc', description: 'unclassified'),
+  vUnspec(id: 'v-unspec', description: 'verb unspecified'),
+  v1(id: 'v1', description: 'Ichidan verb'),
+  v1S(id: 'v1-s', description: 'Ichidan verb - kureru special class'),
+  v2aS(id: 'v2a-s', description: 'Nidan verb with ''u'' ending (archaic)'),
+  v2bK(id: 'v2b-k', description: 'Nidan verb (upper class) with ''bu'' ending (archaic)'),
+  v2bS(id: 'v2b-s', description: 'Nidan verb (lower class) with ''bu'' ending (archaic)'),
+  v2dK(id: 'v2d-k', description: 'Nidan verb (upper class) with ''dzu'' ending (archaic)'),
+  v2dS(id: 'v2d-s', description: 'Nidan verb (lower class) with ''dzu'' ending (archaic)'),
+  v2gK(id: 'v2g-k', description: 'Nidan verb (upper class) with ''gu'' ending (archaic)'),
+  v2gS(id: 'v2g-s', description: 'Nidan verb (lower class) with ''gu'' ending (archaic)'),
+  v2hK(id: 'v2h-k', description: 'Nidan verb (upper class) with ''hu/fu'' ending (archaic)'),
+  v2hS(id: 'v2h-s', description: 'Nidan verb (lower class) with ''hu/fu'' ending (archaic)'),
+  v2kK(id: 'v2k-k', description: 'Nidan verb (upper class) with ''ku'' ending (archaic)'),
+  v2kS(id: 'v2k-s', description: 'Nidan verb (lower class) with ''ku'' ending (archaic)'),
+  v2mK(id: 'v2m-k', description: 'Nidan verb (upper class) with ''mu'' ending (archaic)'),
+  v2mS(id: 'v2m-s', description: 'Nidan verb (lower class) with ''mu'' ending (archaic)'),
+  v2nS(id: 'v2n-s', description: 'Nidan verb (lower class) with ''nu'' ending (archaic)'),
+  v2rK(id: 'v2r-k', description: 'Nidan verb (upper class) with ''ru'' ending (archaic)'),
+  v2rS(id: 'v2r-s', description: 'Nidan verb (lower class) with ''ru'' ending (archaic)'),
+  v2sS(id: 'v2s-s', description: 'Nidan verb (lower class) with ''su'' ending (archaic)'),
+  v2tK(id: 'v2t-k', description: 'Nidan verb (upper class) with ''tsu'' ending (archaic)'),
+  v2tS(id: 'v2t-s', description: 'Nidan verb (lower class) with ''tsu'' ending (archaic)'),
+  v2wS(id: 'v2w-s', description: 'Nidan verb (lower class) with ''u'' ending and ''we'' conjugation (archaic)'),
+  v2yK(id: 'v2y-k', description: 'Nidan verb (upper class) with ''yu'' ending (archaic)'),
+  v2yS(id: 'v2y-s', description: 'Nidan verb (lower class) with ''yu'' ending (archaic)'),
+  v2zS(id: 'v2z-s', description: 'Nidan verb (lower class) with ''zu'' ending (archaic)'),
+  v4b(id: 'v4b', description: 'Yodan verb with ''bu'' ending (archaic)'),
+  v4g(id: 'v4g', description: 'Yodan verb with ''gu'' ending (archaic)'),
+  v4h(id: 'v4h', description: 'Yodan verb with ''hu/fu'' ending (archaic)'),
+  v4k(id: 'v4k', description: 'Yodan verb with ''ku'' ending (archaic)'),
+  v4m(id: 'v4m', description: 'Yodan verb with ''mu'' ending (archaic)'),
+  v4n(id: 'v4n', description: 'Yodan verb with ''nu'' ending (archaic)'),
+  v4r(id: 'v4r', description: 'Yodan verb with ''ru'' ending (archaic)'),
+  v4s(id: 'v4s', description: 'Yodan verb with ''su'' ending (archaic)'),
+  v4t(id: 'v4t', description: 'Yodan verb with ''tsu'' ending (archaic)'),
+  v5aru(id: 'v5aru', description: 'Godan verb - -aru special class'),
+  v5b(id: 'v5b', description: 'Godan verb with ''bu'' ending'),
+  v5g(id: 'v5g', description: 'Godan verb with ''gu'' ending'),
+  v5k(id: 'v5k', description: 'Godan verb with ''ku'' ending'),
+  v5kS(id: 'v5k-s', description: 'Godan verb - Iku/Yuku special class'),
+  v5m(id: 'v5m', description: 'Godan verb with ''mu'' ending'),
+  v5n(id: 'v5n', description: 'Godan verb with ''nu'' ending'),
+  v5r(id: 'v5r', description: 'Godan verb with ''ru'' ending'),
+  v5rI(id: 'v5r-i', description: 'Godan verb with ''ru'' ending (irregular verb)'),
+  v5s(id: 'v5s', description: 'Godan verb with ''su'' ending'),
+  v5t(id: 'v5t', description: 'Godan verb with ''tsu'' ending'),
+  v5u(id: 'v5u', description: 'Godan verb with ''u'' ending'),
+  v5uS(id: 'v5u-s', description: 'Godan verb with ''u'' ending (special class)'),
+  v5uru(id: 'v5uru', description: 'Godan verb - Uru old class verb (old form of Eru)'),
+  vi(id: 'vi', description: 'intransitive verb'),
+  vk(id: 'vk', description: 'Kuru verb - special class'),
+  vn(id: 'vn', description: 'irregular nu verb'),
+  vr(id: 'vr', description: 'irregular ru verb, plain form ends with -ri'),
+  vs(id: 'vs', description: 'noun or participle which takes the aux. verb suru'),
+  vsC(id: 'vs-c', description: 'suru verb - precursor to the modern suru'),
+  vsI(id: 'vs-i', description: 'suru verb - included'),
+  vsS(id: 'vs-s', description: 'suru verb - special class'),
+  vt(id: 'vt', description: 'transitive verb'),
+  vz(id: 'vz', description: 'Ichidan verb - zuru verb (alternative form of -jiru verbs)');
+
+  final String id;
+  final String description;
+
+  const JMdictPOS({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictPOS fromId(String id) =>
+      JMdictPOS.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictPOS fromJson(Map<String, Object?> json) =>
+      JMdictPOS.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/jmdict_reading_info.dart

@@ -0,0 +1,39 @@
+/// Reading info tags from JMdict
+///
+/// See https://www.edrdg.org/jmwsgi/edhelp.py#kw_rinf
+enum JMdictReadingInfo {
+  gikun(
+    id: 'gikun',
+    description:
+        'gikun (meaning as reading) or jukujikun (special kanji reading)',
+  ),
+  ik(id: 'ik', description: 'word containing irregular kana usage'),
+  ok(id: 'ok', description: 'out-dated or obsolete kana usage'),
+  rk(id: 'rk', description: 'rarely used kana form'),
+  sk(id: 'sk', description: 'search-only kana form');
+
+  final String id;
+  final String description;
+
+  const JMdictReadingInfo({
+    required this.id,
+    required this.description,
+  });
+
+  static JMdictReadingInfo fromId(String id) =>
+      JMdictReadingInfo.values.firstWhere(
+        (e) => e.id == id,
+        orElse: () => throw Exception('Unknown id: $id'),
+      );
+
+  Map<String, Object?> toJson() => {
+        'id': id,
+        'description': description,
+      };
+
+  static JMdictReadingInfo fromJson(Map<String, Object?> json) =>
+      JMdictReadingInfo.values.firstWhere(
+        (e) => e.id == json['id'],
+        orElse: () => throw Exception('Unknown id: ${json['id']}'),
+      );
+}

lib/models/jmdict/word_search_result.dart

@@ -1,5 +0,0 @@
-class WordSearchResult {
-  // TODO: implement me
-
-  Map<String, dynamic> toJson() => {};
-}

lib/models/word_search/word_search_result.dart

@@ -0,0 +1,25 @@
+import 'package:jadb/models/word_search/word_search_ruby.dart';
+import 'package:jadb/models/word_search/word_search_sense.dart';
+import 'package:jadb/models/word_search/word_search_sources.dart';
+
+class WordSearchResult {
+  // TODO: implement me
+
+  final int entryId;
+  final List<WordSearchRuby> japanese;
+  final List<WordSearchSense> senses;
+  final WordSearchSources sources;
+
+  const WordSearchResult({
+    required this.entryId,
+    required this.japanese,
+    required this.senses,
+    required this.sources,
+  });
+
+  Map<String, dynamic> toJson() => {
+        'entryId': entryId,
+        'japanese': japanese.map((e) => e.toJson()).toList(),
+        'senses': senses.map((e) => e.toJson()).toList(),
+      };
+}

lib/models/word_search/word_search_ruby.dart

@@ -0,0 +1,25 @@
+/// A pair of base and optional furigana.
+class WordSearchRuby {
+  /// Base part. Could be a kanji or a reading.
+  String base;
+
+  /// Furigana, if applicable.
+  String? furigana;
+
+  WordSearchRuby({
+    required this.base,
+    this.furigana,
+  });
+
+  Map<String, dynamic> toJson() => {
+        'base': base,
+        'furigana': furigana,
+      };
+
+  factory WordSearchRuby.fromJson(Map<String, dynamic> json) {
+    return WordSearchRuby(
+      base: json['base'] as String,
+      furigana: json['furigana'] as String?,
+    );
+  }
+}

lib/models/word_search/word_search_sense.dart

@@ -0,0 +1,100 @@
+import 'package:jadb/models/jmdict/jmdict_dialect.dart';
+import 'package:jadb/models/jmdict/jmdict_field.dart';
+import 'package:jadb/models/jmdict/jmdict_misc.dart';
+import 'package:jadb/models/jmdict/jmdict_pos.dart';
+import 'package:jadb/models/word_search/word_search_xref_entry.dart';
+
+class WordSearchSense {
+  /// The meaning(s) of the word.
+  final List<String> englishDefinitions;
+
+  /// Type of word (Noun, Verb, etc.).
+  final List<JMdictPOS> partsOfSpeech;
+
+  /// Relevant words (might include synonyms).
+  final List<WordSearchXrefEntry> seeAlso;
+
+  /// Words with opposite meaning.
+  final List<WordSearchXrefEntry> antonyms;
+
+  /// Restrictions on which of the readings of the parent entry this sense applies to.
+  final List<String> restrictedToReading;
+
+  /// Restrictions on which of the kanji of the parent entry this sense applies to.
+  final List<String> restrictedToKanji;
+
+  /// Tags for which domains or fields of expertise that this sense is relevant to.
+  final List<JMdictField> fields;
+
+  /// Tags for which dialects this sense is used in.
+  final List<JMdictDialect> dialects;
+
+  /// Tags for miscellaneous information.
+  final List<JMdictMisc> misc;
+
+  /// Extra information about the sense.
+  final List<String> info;
+
+  // TODO: there is a lot more info to collect in the languageSource data
+
+  /// Information about the the origin of the word, if loaned from another language.
+  final List<String> languageSource;
+
+  // TODO: add example sentencesa
+
+  const WordSearchSense({
+    required this.englishDefinitions,
+    required this.partsOfSpeech,
+    required this.seeAlso,
+    required this.antonyms,
+    required this.restrictedToReading,
+    required this.restrictedToKanji,
+    required this.fields,
+    required this.dialects,
+    required this.misc,
+    required this.info,
+    required this.languageSource,
+  });
+  
+  bool get isEmpty => englishDefinitions.isEmpty &&
+      partsOfSpeech.isEmpty &&
+      seeAlso.isEmpty &&
+      antonyms.isEmpty &&
+      restrictedToReading.isEmpty &&
+      restrictedToKanji.isEmpty &&
+      fields.isEmpty &&
+      dialects.isEmpty &&
+      misc.isEmpty &&
+      info.isEmpty &&
+      languageSource.isEmpty;
+
+  Map<String, dynamic> toJson() => {
+        'englishDefinitions': englishDefinitions,
+        'partsOfSpeech': partsOfSpeech.map((e) => e.toJson()).toList(),
+        'seeAlso': seeAlso.map((e) => e.toJson()).toList(),
+        'antonyms': antonyms.map((e) => e.toJson()).toList(),
+        'restrictedToReading': restrictedToReading,
+        'restrictedToKanji': restrictedToKanji,
+        'fields': fields.map((e) => e.toJson()).toList(),
+        'dialects': dialects.map((e) => e.toJson()).toList(),
+        'misc': misc.map((e) => e.toJson()).toList(),
+        'info': info,
+        'languageSource': languageSource,
+      };
+  
+  String prettyPrint() {
+    return 'WordSearchSense{'
+        'englishDefinitions: $englishDefinitions, '
+        'partsOfSpeech: $partsOfSpeech, '
+        'seeAlso: $seeAlso, '
+        'antonyms: $antonyms, '
+        'restrictedToReading: $restrictedToReading, '
+        'restrictedToKanji: $restrictedToKanji, '
+        'fields: $fields, '
+        'dialects: $dialects, '
+        'misc: $misc, '
+        'info: $info, '
+        'languageSource: $languageSource'
+        '}';
+  }
+}

lib/models/word_search/word_search_sources.dart

@@ -0,0 +1,21 @@
+class WordSearchSources {
+  final bool jmdict;
+  final bool jmnedict;
+
+  const WordSearchSources({
+    this.jmdict = true,
+    this.jmnedict = false,
+  });
+
+  Map<String, Object?> get sqlValue => {
+        'jmdict': jmdict,
+        'jmnedict': jmnedict,
+      };
+  
+  factory WordSearchSources.fromJson(Map<String, dynamic> json) {
+    return WordSearchSources(
+      jmdict: json['jmdict'] as bool? ?? true,
+      jmnedict: json['jmnedict'] as bool? ?? false,
+    );
+  }
+}

lib/models/word_search/word_search_xref_entry.dart

@@ -0,0 +1,20 @@
+class WordSearchXrefEntry {
+  final int entryId;
+  final bool ambiguous;
+
+  const WordSearchXrefEntry({
+    required this.entryId,
+    required this.ambiguous,
+  });
+
+  Map<String, dynamic> toJson() => {
+        'entryId': entryId,
+        'ambiguous': ambiguous,
+      };
+
+  factory WordSearchXrefEntry.fromJson(Map<String, dynamic> json) =>
+      WordSearchXrefEntry(
+        entryId: json['entryId'] as int,
+        ambiguous: json['ambiguous'] as bool,
+      );
+}

lib/search.dart

@@ -1,5 +1,5 @@
-import 'package:jadb/models/jmdict/word_search_result.dart';
-import 'package:jadb/models/kanjidic/kanji_search_result.dart';
+import 'package:jadb/models/word_search/word_search_result.dart';
+import 'package:jadb/models/kanji_search/kanji_search_result.dart';
 import 'package:jadb/models/radkfile/radicals_search_result.dart';
 import 'package:jadb/search/jmdict.dart';
 

lib/search/jmdict.dart

@@ -1,4 +1,12 @@
-import 'package:jadb/models/jmdict/word_search_result.dart';
+import 'package:collection/collection.dart';
+import 'package:jadb/models/jmdict/jmdict_dialect.dart';
+import 'package:jadb/models/jmdict/jmdict_field.dart';
+import 'package:jadb/models/jmdict/jmdict_misc.dart';
+import 'package:jadb/models/jmdict/jmdict_pos.dart';
+import 'package:jadb/models/word_search/word_search_result.dart';
+import 'package:jadb/models/word_search/word_search_ruby.dart';
+import 'package:jadb/models/word_search/word_search_sense.dart';
+import 'package:jadb/models/word_search/word_search_xref_entry.dart';
 import 'package:sqflite_common/sqlite_api.dart';
 
 // TODO: Support globs
@@ -9,6 +17,10 @@ import 'package:sqflite_common/sqlite_api.dart';
 
 // TODO: Support mixing kana and romaji
 
+String _escapeStringValue(String value) {
+  return "'" + value.replaceAll("'", "''") + "'";
+}
+
 Future<List<WordSearchResult>?> searchWordWithDbConnection(
   DatabaseExecutor connection,
   String word, {
@@ -18,176 +30,176 @@ Future<List<WordSearchResult>?> searchWordWithDbConnection(
     return null;
   }
 
-  late final List<int> matches;
+  late final List<int> entryIds;
   if (isKana) {
-    matches = (await connection.query(
+    entryIds = (await connection.query(
       'JMdict_EntryByKana',
       where: 'kana LIKE ?',
-      whereArgs: ['%$word%'],
+      whereArgs: ['$word%'],
     ))
         .map((row) => row['entryId'] as int)
         .toList();
   } else {
-    matches = (await connection.query(
+    entryIds = (await connection.query(
       'JMdict_EntryByEnglish',
       where: 'english LIKE ?',
-      whereArgs: ['%$word%'],
+      whereArgs: ['$word%'],
     ))
         .map((row) => row['entryId'] as int)
         .toList();
   }
 
-  if (matches.isEmpty) {
+  if (entryIds.isEmpty) {
     return [];
   }
 
-  late final List<int> senseIds;
-  final Future<List<int>> senseIds_query = connection
-      .query(
-        'JMdict_Sense',
-        where: 'entryId IN (${matches.join(',')})',
-      )
-      .then((rows) => rows.map((row) => row['id'] as int).toList());
+  late final List<Map<String, Object?>> senses;
+  final Future<List<Map<String, Object?>>> senses_query = connection.query(
+    'JMdict_Sense',
+    where: 'entryId IN (${entryIds.join(',')})',
+  );
 
   late final List<Map<String, Object?>> readingElements;
   final Future<List<Map<String, Object?>>> readingElements_query =
       connection.query(
     'JMdict_ReadingElement',
-    where: 'entryId IN (${matches.join(',')})',
+    where: 'entryId IN (${entryIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> kanjiElements;
   final Future<List<Map<String, Object?>>> kanjiElements_query =
       connection.query(
     'JMdict_KanjiElement',
-    where: 'entryId IN (${matches.join(',')})',
+    where: 'entryId IN (${entryIds.join(',')})',
   );
 
   await Future.wait([
-    senseIds_query.then((value) => senseIds = value),
+    senses_query.then((value) => senses = value),
     readingElements_query.then((value) => readingElements = value),
     kanjiElements_query.then((value) => kanjiElements = value),
   ]);
 
-  print(senseIds);
-  print(readingElements);
-  print(kanjiElements);
-
   // Sense queries
 
+  final senseIds = senses.map((element) => element['id'] as int).toList();
+
   late final List<Map<String, Object?>> senseAntonyms;
   final Future<List<Map<String, Object?>>> senseAntonyms_query =
       connection.query(
     'JMdict_SenseAntonym',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseDialects;
   final Future<List<Map<String, Object?>>> senseDialects_query =
       connection.query(
     'JMdict_SenseDialect',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseFields;
   final Future<List<Map<String, Object?>>> senseFields_query = connection.query(
     'JMdict_SenseField',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseGlossaries;
   final Future<List<Map<String, Object?>>> senseGlossaries_query =
       connection.query(
     'JMdict_SenseGlossary',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseInfos;
   final Future<List<Map<String, Object?>>> senseInfos_query = connection.query(
     'JMdict_SenseInfo',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseLanguageSources;
   final Future<List<Map<String, Object?>>> senseLanguageSources_query =
       connection.query(
     'JMdict_SenseLanguageSource',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseMiscs;
   final Future<List<Map<String, Object?>>> senseMiscs_query = connection.query(
     'JMdict_SenseMisc',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> sensePOSs;
   final Future<List<Map<String, Object?>>> sensePOSs_query = connection.query(
     'JMdict_SensePOS',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseRestrictedToKanjis;
   final Future<List<Map<String, Object?>>> senseRestrictedToKanjis_query =
       connection.query(
     'JMdict_SenseRestrictedToKanji',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseRestrictedToReadings;
   final Future<List<Map<String, Object?>>> senseRestrictedToReadings_query =
       connection.query(
     'JMdict_SenseRestrictedToReading',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> senseSeeAlsos;
   final Future<List<Map<String, Object?>>> senseSeeAlsos_query =
       connection.query(
     'JMdict_SenseSeeAlso',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> exampleSentences;
   final Future<List<Map<String, Object?>>> exampleSentences_query =
       connection.query(
     'JMdict_ExampleSentence',
-    where: 'entryId IN (${senseIds.join(',')})',
+    where: 'senseId IN (${senseIds.join(',')})',
   );
 
   // Reading queries
 
   final readingIds = readingElements
-      .map((element) =>
-          (element['entryId'] as int, element['reading'] as String))
+      .map((element) => (
+            element['entryId'] as int,
+            _escapeStringValue(element['reading'] as String)
+          ))
       .toList();
 
   late final List<Map<String, Object?>> readingElementInfos;
   final Future<List<Map<String, Object?>>> readingElementInfos_query =
       connection.query(
     'JMdict_ReadingElementInfo',
-    where: 'entryId IN (${readingIds.join(',')})',
+    where: '(entryId, reading) IN (${readingIds.join(',')})',
   );
 
   late final List<Map<String, Object?>> readingElementRestrictions;
   final Future<List<Map<String, Object?>>> readingElementRestrictions_query =
       connection.query(
     'JMdict_ReadingElementRestriction',
-    where: 'entryId IN (${readingIds.join(',')})',
+    where: '(entryId, reading) IN (${readingIds.join(',')})',
   );
 
   // Kanji queries
 
   final kanjiIds = kanjiElements
-      .map((element) =>
-          (element['entryId'] as int, element['reading'] as String))
+      .map((element) => (
+            element['entryId'] as int,
+            _escapeStringValue(element['reading'] as String)
+          ))
       .toList();
 
   late final List<Map<String, Object?>> kanjiElementInfos;
   final Future<List<Map<String, Object?>>> kanjiElementInfos_query =
       connection.query(
     'JMdict_KanjiElementInfo',
-    where: 'entryId IN (${kanjiIds.join(',')})',
+    where: '(entryId, reading) IN (${kanjiIds.join(',')})',
   );
 
   await Future.wait([
@@ -211,5 +223,251 @@ Future<List<WordSearchResult>?> searchWordWithDbConnection(
     kanjiElementInfos_query.then((value) => kanjiElementInfos = value),
   ]);
 
-  throw UnimplementedError();
+  return _regroupWordSearchResults(
+    entryIds: entryIds,
+    readingElements: readingElements,
+    kanjiElements: kanjiElements,
+    senses: senses,
+    senseAntonyms: senseAntonyms,
+    senseDialects: senseDialects,
+    senseFields: senseFields,
+    senseGlossaries: senseGlossaries,
+    senseInfos: senseInfos,
+    senseLanguageSources: senseLanguageSources,
+    senseMiscs: senseMiscs,
+    sensePOSs: sensePOSs,
+    senseRestrictedToKanjis: senseRestrictedToKanjis,
+    senseRestrictedToReadings: senseRestrictedToReadings,
+    senseSeeAlsos: senseSeeAlsos,
+    exampleSentences: exampleSentences,
+    readingElementInfos: readingElementInfos,
+    readingElementRestrictions: readingElementRestrictions,
+    kanjiElementInfos: kanjiElementInfos,
+  );
+}
+
+List<WordSearchResult> _regroupWordSearchResults({
+  required List<int> entryIds,
+  required List<Map<String, Object?>> readingElements,
+  required List<Map<String, Object?>> kanjiElements,
+  required List<Map<String, Object?>> senses,
+  required List<Map<String, Object?>> senseAntonyms,
+  required List<Map<String, Object?>> senseDialects,
+  required List<Map<String, Object?>> senseFields,
+  required List<Map<String, Object?>> senseGlossaries,
+  required List<Map<String, Object?>> senseInfos,
+  required List<Map<String, Object?>> senseLanguageSources,
+  required List<Map<String, Object?>> senseMiscs,
+  required List<Map<String, Object?>> sensePOSs,
+  required List<Map<String, Object?>> senseRestrictedToKanjis,
+  required List<Map<String, Object?>> senseRestrictedToReadings,
+  required List<Map<String, Object?>> senseSeeAlsos,
+  required List<Map<String, Object?>> exampleSentences,
+  required List<Map<String, Object?>> readingElementInfos,
+  required List<Map<String, Object?>> readingElementRestrictions,
+  required List<Map<String, Object?>> kanjiElementInfos,
+}) {
+  final List<WordSearchResult> results = [];
+
+  for (final entryId in entryIds) {
+    final List<Map<String, Object?>> entryReadingElements = readingElements
+        .where((element) => element['entryId'] == entryId)
+        .toList();
+
+    final List<Map<String, Object?>> entryKanjiElements = kanjiElements
+        .where((element) => element['entryId'] == entryId)
+        .toList();
+
+    final List<Map<String, Object?>> entrySenses =
+        senses.where((element) => element['entryId'] == entryId).toList();
+
+    final List<WordSearchRuby> entryReadingElementsGrouped = _regroup_words(
+      entryId: entryId,
+      readingElements: entryReadingElements,
+      kanjiElements: entryKanjiElements,
+      readingElementInfos: readingElementInfos,
+      readingElementRestrictions: readingElementRestrictions,
+      kanjiElementInfos: kanjiElementInfos,
+    );
+
+    final List<WordSearchSense> entrySensesGrouped = _regroup_senses(
+      senses: entrySenses,
+      senseAntonyms: senseAntonyms,
+      senseDialects: senseDialects,
+      senseFields: senseFields,
+      senseGlossaries: senseGlossaries,
+      senseInfos: senseInfos,
+      senseLanguageSources: senseLanguageSources,
+      senseMiscs: senseMiscs,
+      sensePOSs: sensePOSs,
+      senseRestrictedToKanjis: senseRestrictedToKanjis,
+      senseRestrictedToReadings: senseRestrictedToReadings,
+      senseSeeAlsos: senseSeeAlsos,
+      exampleSentences: exampleSentences,
+    );
+
+    print("Entry ID: $entryId");
+    print("Words:");
+    for (final ruby in entryReadingElementsGrouped) {
+      print("${ruby.base}[${ruby.furigana ?? ""}]");
+    }
+    print("Senses:");
+    for (final sense in entrySensesGrouped) {
+      print(sense.prettyPrint());
+    }
+  }
+
+  return results;
+}
+
+List<WordSearchRuby> _regroup_words({
+  required int entryId,
+  required List<Map<String, Object?>> kanjiElements,
+  required List<Map<String, Object?>> kanjiElementInfos,
+  required List<Map<String, Object?>> readingElements,
+  required List<Map<String, Object?>> readingElementInfos,
+  required List<Map<String, Object?>> readingElementRestrictions,
+}) {
+  final List<WordSearchRuby> result = [];
+
+  final kanjiElements_ =
+      kanjiElements.where((element) => element['entryId'] == entryId).toList();
+
+  final readingElements_ = readingElements
+      .where((element) => element['entryId'] == entryId)
+      .toList();
+
+  final readingElementRestrictions_ = readingElementRestrictions
+      .where((element) => element['entryId'] == entryId)
+      .toList();
+
+  for (final readingElement in readingElements_) {
+    for (final kanjiElement in kanjiElements_) {
+      final kanji = kanjiElement['reading'] as String;
+      final reading = readingElement['reading'] as String;
+
+      final doesNotMatchKanji = readingElement['doesNotMatchKanji'] == 1;
+      if (doesNotMatchKanji) {
+        continue;
+      }
+
+      final restrictions = readingElementRestrictions_
+          .where((element) => element['reading'] == reading)
+          .toList();
+
+      if (restrictions.isNotEmpty &&
+          !restrictions.any((element) => element['restriction'] == kanji)) {
+        continue;
+      }
+
+      final ruby = WordSearchRuby(
+        base: kanji,
+        furigana: reading,
+      );
+      result.add(ruby);
+    }
+  }
+
+  for (final readingElement
+      in readingElements_.where((e) => e['doesNotMatchKanji'] == 1)) {
+    final reading = readingElement['reading'] as String;
+    final ruby = WordSearchRuby(
+      base: reading,
+    );
+    result.add(ruby);
+  }
+
+  return result;
+}
+
+List<WordSearchSense> _regroup_senses({
+  required List<Map<String, Object?>> senses,
+  required List<Map<String, Object?>> senseAntonyms,
+  required List<Map<String, Object?>> senseDialects,
+  required List<Map<String, Object?>> senseFields,
+  required List<Map<String, Object?>> senseGlossaries,
+  required List<Map<String, Object?>> senseInfos,
+  required List<Map<String, Object?>> senseLanguageSources,
+  required List<Map<String, Object?>> senseMiscs,
+  required List<Map<String, Object?>> sensePOSs,
+  required List<Map<String, Object?>> senseRestrictedToKanjis,
+  required List<Map<String, Object?>> senseRestrictedToReadings,
+  required List<Map<String, Object?>> senseSeeAlsos,
+  required List<Map<String, Object?>> exampleSentences,
+}) {
+  final groupedSenseAntonyms =
+      senseAntonyms.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseDialects =
+      senseDialects.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseFields =
+      senseFields.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseGlossaries =
+      senseGlossaries.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseInfos =
+      senseInfos.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseLanguageSources =
+      senseLanguageSources.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseMiscs =
+      senseMiscs.groupListsBy((element) => element['senseId'] as int);
+  final groupedSensePOSs =
+      sensePOSs.groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseRestrictedToKanjis = senseRestrictedToKanjis
+      .groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseRestrictedToReadings = senseRestrictedToReadings
+      .groupListsBy((element) => element['senseId'] as int);
+  final groupedSenseSeeAlsos =
+      senseSeeAlsos.groupListsBy((element) => element['senseId'] as int);
+
+  final List<WordSearchSense> result = [];
+  for (final sense in senses) {
+    final int senseId = sense['id'] as int;
+
+    final antonyms = groupedSenseAntonyms[senseId] ?? [];
+    final dialects = groupedSenseDialects[senseId] ?? [];
+    final fields = groupedSenseFields[senseId] ?? [];
+    final glossaries = groupedSenseGlossaries[senseId] ?? [];
+    final infos = groupedSenseInfos[senseId] ?? [];
+    final languageSources = groupedSenseLanguageSources[senseId] ?? [];
+    final miscs = groupedSenseMiscs[senseId] ?? [];
+    final pos = groupedSensePOSs[senseId] ?? [];
+    final restrictedToKanjis = groupedSenseRestrictedToKanjis[senseId] ?? [];
+    final restrictedToReadings =
+        groupedSenseRestrictedToReadings[senseId] ?? [];
+    final seeAlsos = groupedSenseSeeAlsos[senseId] ?? [];
+
+    final resultSense = WordSearchSense(
+      englishDefinitions: glossaries.map((e) => e['phrase'] as String).toList(),
+      partsOfSpeech:
+          pos.map((e) => JMdictPOS.fromId(e['pos'] as String)).toList(),
+      seeAlso: seeAlsos
+          .map((e) => WordSearchXrefEntry(
+                entryId: e['xrefEntryId'] as int,
+                ambiguous: e['ambiguous'] == 1,
+              ))
+          .toList(),
+      antonyms: antonyms
+          .map((e) => WordSearchXrefEntry(
+                entryId: e['xrefEntryId'] as int,
+                ambiguous: e['ambiguous'] == 1,
+              ))
+          .toList(),
+      restrictedToReading:
+          restrictedToReadings.map((e) => e['reading'] as String).toList(),
+      restrictedToKanji:
+          restrictedToKanjis.map((e) => e['kanji'] as String).toList(),
+      fields:
+          fields.map((e) => JMdictField.fromId(e['field'] as String)).toList(),
+      dialects: dialects
+          .map((e) => JMdictDialect.fromId(e['dialect'] as String))
+          .toList(),
+      misc: miscs.map((e) => JMdictMisc.fromId(e['misc'] as String)).toList(),
+      info: infos.map((e) => e['info'] as String).toList(),
+      languageSource:
+          languageSources.map((e) => e['language'] as String).toList(),
+    );
+
+    result.add(resultSense);
+  }
+
+  return result;
 }

lib/search/kanji.dart

@@ -1,4 +1,4 @@
-import 'package:jadb/models/kanjidic/kanji_search_result.dart';
+import 'package:jadb/models/kanji_search/kanji_search_result.dart';
 import 'package:sqflite_common/sqflite.dart';
 
 Future<KanjiSearchResult?> searchKanjiWithDbConnection(

migrations/0002_JMDict_insert_constants.sql

@@ -274,18 +274,22 @@ INSERT INTO "JMdict_InfoPOS"("id", "description") VALUES
   ('vt', 'transitive verb'),
   ('vz', 'Ichidan verb - zuru verb (alternative form of -jiru verbs)');
 
+-- NOTE: several of these are not used directly in the XML export, but is implicitly
+--       used by whether the reading is marked as kun/on and nanori tags, etc.
+
 INSERT INTO "JMdict_InfoReading"("id", "description") VALUES
   ('gikun', 'gikun (meaning as reading) or jukujikun (special kanji reading)'),
-  ('go', 'on-yomi, go'),
+  -- ('go', 'on-yomi, go'),
   ('ik', 'word containing irregular kana usage'),
-  ('jouyou', 'approved reading for jouyou kanji'),
-  ('kan', 'on-yomi, kan'),
-  ('kanyou', 'on-yomi, kan''you'),
-  ('kun', 'kun-yomi'),
-  ('name', 'reading used only in names (nanori)'),
+  -- ('jouyou', 'approved reading for jouyou kanji'),
+  -- ('kan', 'on-yomi, kan'),
+  -- ('kanyou', 'on-yomi, kan''you'),
+  -- ('kun', 'kun-yomi'),
+  -- ('name', 'reading used only in names (nanori)'),
   ('ok', 'out-dated or obsolete kana usage'),
-  ('on', 'on-yomi'),
-  ('rad', 'reading used as name of radical'),
+  -- ('on', 'on-yomi'),
+  -- ('rad', 'reading used as name of radical'),
   ('rk', 'rarely used kana form'),
-  ('sk', 'search-only kana form'),
-  ('tou', 'on-yomi, tou');
+  ('sk', 'search-only kana form')
+  -- ('tou', 'on-yomi, tou')
+  ;