WIP

Add datasource versions to database
benchmark: create separate benchmarks for jp and en search
2026-04-13 21:10:56 +09:00 · 2026-04-13 21:00:39 +09:00 · 2026-04-13 20:24:26 +09:00 · 2026-04-13 20:24:25 +09:00 · 2026-04-13 20:24:25 +09:00 · 2026-04-13 19:42:11 +09:00
47 changed files with 704 additions and 8838 deletions
--- a/.gitea/workflows/build-and-test.yml
+++ b/.gitea/workflows/build-and-test.yml
@@ -24,11 +24,7 @@ jobs:
          build-users-group =

    - name: Update database inputs
-      run: |
-        nix flake update jmdict-src
-        nix flake update jmdict-with-examples-src
-        nix flake update radkfile-src
-        nix flake update kanjidic2-src
+      run: nix flake update datasources

    - name: Build database
      run: nix build .#database -L
@@ -44,7 +40,7 @@ jobs:
        compression: 0

    - name: Print database statistics
-      run: nix develop .# --command sqlite3_analyzer result/jadb.sqlite
+      run: nix develop .#sqlite-debugging --command sqlite3_analyzer result/jadb.sqlite

    # TODO: Defer failure of tests until after the coverage report is generated and uploaded.
    - name: Run tests
--- a/README.md
+++ b/README.md
@@ -18,26 +18,4 @@ Note that while the license for the code is MIT, the data has various licenses.
 | **Tanos JLPT levels:** | https://www.tanos.co.uk/jlpt/ |
 | **Kangxi Radicals:**   | https://ctext.org/kangxi-zidian |

-## Implementation details
-
-### Word search
-
-The word search procedure is currently split into 3 parts:
-
-1. **Entry ID query**:
-
-Use a complex query with various scoring factors to try to get list of
-database ids pointing at dictionary entries, sorted by how likely we think this
-word is the word that the caller is looking for. The output here is a `List<int>`
-
-2. **Data Query**:
-
-Takes the entry id list from the last search, and performs all queries needed to retrieve
-all the dictionary data for those IDs. The result is a struct with a bunch of flattened lists
-with data for all the dictionary entries. These lists are sorted by the order that the ids
-were provided.
-
-3. **Regrouping**:
-
-Takes the flattened data, and regroups the items into structs with a more "hierarchical" structure.
-All data tagged with the same ID will end up in the same struct. Returns a list of these structs.
+See [docs/overview.md](./docs/overview.md) for notes and implementation details.
--- a/benchmark/benchmark.dart
+++ b/benchmark/benchmark.dart
@@ -1,5 +1,7 @@
-import './search/word_search.dart';
+import './search/english_word_search.dart';
+import './search/japanese_word_search.dart';

 Future<void> main() async {
-  await WordSearchBenchmark.main();
+  await EnglishWordSearchBenchmark.main();
+  await JapaneseWordSearchBenchmark.main();
 }
--- a/benchmark/search/english_word_search.dart
+++ b/benchmark/search/english_word_search.dart
@@ -4,23 +4,23 @@ import 'package:sqflite_common/sqlite_api.dart';

 import '../../test/search/setup_database_connection.dart';

-class WordSearchBenchmark extends AsyncBenchmarkBase {
+class EnglishWordSearchBenchmark extends AsyncBenchmarkBase {
  Database? connection;

  static final List<String> searchTerms = [
    'kana',
    'kanji',
-    'kawaii',
+    'cute',
    'sushi',
    'ramen',
  ];

-  WordSearchBenchmark() : super('WordSearchBenchmark');
+  EnglishWordSearchBenchmark() : super('EnglishWordSearchBenchmark');

  static Future<void> main() async {
-    print('Running WordSearchBenchmark...');
-    await WordSearchBenchmark().report();
-    print('Finished WordSearchBenchmark');
+    print('Running EnglishWordSearchBenchmark...');
+    await EnglishWordSearchBenchmark().report();
+    print('Finished EnglishWordSearchBenchmark');
  }

  @override
@@ -31,7 +31,11 @@ class WordSearchBenchmark extends AsyncBenchmarkBase {
  @override
  Future<void> run() async {
    for (final term in searchTerms) {
-      await connection!.jadbSearchWord(term);
+      final result = await connection!.jadbSearchWord(term);
+      assert(
+        result?.isNotEmpty ?? false,
+        'Expected search results for term "$term"',
+      );
    }
  }

--- a/benchmark/search/japanese_word_search.dart
+++ b/benchmark/search/japanese_word_search.dart
@@ -0,0 +1,49 @@
+import 'package:benchmark_harness/benchmark_harness.dart';
+import 'package:jadb/search.dart';
+import 'package:sqflite_common/sqlite_api.dart';
+
+import '../../test/search/setup_database_connection.dart';
+
+class JapaneseWordSearchBenchmark extends AsyncBenchmarkBase {
+  Database? connection;
+
+  static final List<String> searchTerms = [
+    '仮名',
+    '漢字',
+    'かわいい',
+    'すし',
+    'ラメン',
+  ];
+
+  JapaneseWordSearchBenchmark() : super('JapaneseWordSearchBenchmark');
+
+  static Future<void> main() async {
+    print('Running JapaneseWordSearchBenchmark...');
+    await JapaneseWordSearchBenchmark().report();
+    print('Finished JapaneseWordSearchBenchmark');
+  }
+
+  @override
+  Future<void> setup() async {
+    connection = await setupDatabaseConnection();
+  }
+
+  @override
+  Future<void> run() async {
+    for (final term in searchTerms) {
+      final result = await connection!.jadbSearchWord(term);
+      assert(
+        result?.isNotEmpty ?? false,
+        'Expected search results for term "$term"',
+      );
+    }
+  }
+
+  @override
+  Future<void> teardown() async {
+    await connection?.close();
+  }
+
+  // @override
+  // Future<void> exercise() => run();
+}
--- a/data/tanos-jlpt/n1.csv
+++ b/data/tanos-jlpt/n1.csv
--- a/data/tanos-jlpt/n2.csv
+++ b/data/tanos-jlpt/n2.csv
--- a/data/tanos-jlpt/n3.csv
+++ b/data/tanos-jlpt/n3.csv
--- a/data/tanos-jlpt/n4.csv
+++ b/data/tanos-jlpt/n4.csv
@@ -1,582 +0,0 @@
-,あ,Ah
-,ああ,like that
-間,あいだ,a space
-合う,あう,to match
-,あかちゃん,baby
-上る,あがる,to rise
-赤ん坊,あかんぼう,baby
-空く,あく,"to open, to become empty"
-,あげる,to give
-浅い,あさい,"shallow, superficial"
-味,あじ,flavour
-明日,あす・あした,tomorrow
-遊び,あそび,play
-集る,あつまる,to gather
-集める,あつめる,to collect something
-謝る,あやまる,to apologize
-安心,あんしん,relief
-安全,あんぜん,safety
-,あんな,such
-以下,いか,less than
-以外,いがい,with the exception of
-医学,いがく,medical science
-生きる,いきる,to live
-意見,いけん,opinion
-石,いし,stone
-,いじめる,to tease
-以上,いじょう,"more than, this is all"
-急ぐ,いそぐ,to hurry
-致す,いたす,(humble) to do
-一度,いちど,once
-一生懸命,いっしょうけんめい,with utmost effort
-,いっぱい,full
-糸,いと,thread
-以内,いない,within
-田舎,いなか,countryside
-祈る,いのる,to pray
-,いらっしゃる,"(respectful) to be, to come or to go"
-植える,うえる,"to plant, to grow"
-受付,うけつけ,receipt
-受ける,うける,to take a lesson or test
-動く,うごく,to move
-,うち,within
-打つ,うつ,to hit
-美しい,うつくしい,beautiful
-写す,うつす,to copy or photograph
-移る,うつる,to move house or transfer
-腕,うで,arm
-裏,うら,reverse side
-売り場,うりば,place where things are sold
-,うん,(informal) yes
-運転手,うんてんしゅ,driver
-枝,えだ,"branch, twig"
-選ぶ,えらぶ,to choose
-遠慮,えんりょ・する,"to be reserved, to be restrained"
-,おいでになる,(respectful) to be
-お祝い,おいわい,congratulation
-,おかげ,"owing to, thanks to"
-,おかしい,strange or funny
-億,おく,one hundred million
-屋上,おくじょう,rooftop
-贈り物,おくりもの,gift
-送る,おくる,to send
-遅れる,おくれる,to be late
-起す,おこす,to wake
-行う,おこなう,to do
-怒る,おこる,"to get angry, to be angry"
-押し入れ,おしいれ,closet
-お嬢さん,おじょうさん,young lady
-お宅,おたく,(polite) your house
-落る,おちる,to fall or drop
-,おっしゃる,(respectful) to say
-夫,おっと,husband
-,おつり,"change from purchase, balance"
-音,おと,"sound, note"
-落す,おとす,to drop
-踊り,おどり,a dance
-踊る,おどる,to dance
-驚く,おどろく,to be surprised
-お祭り,おまつり,festival
-お見舞い,おみまい,"calling on someone who is ill, enquiry"
-お土産,おみやげ,souvenir
-思い出す,おもいだす,to remember
-思う,おもう,"to think, to feel"
-,おもちゃ,toy
-表,おもて,the front
-親,おや,parents
-泳ぎ方,およぎかた,way of swimming
-下りる,おりる,"to get off, to descend"
-折る,おる,to break or to fold
-お礼,おれい,expression of gratitude
-折れる,おれる,to break or be folded
-終わり,おわり,the end
-海岸,かいがん,coast
-会議,かいぎ,meeting
-会議室,かいぎしつ,meeting room
-会場,かいじょう,assembly hall or meeting place
-会話,かいわ,conversation
-帰り,かえり,return
-変える,かえる,to change
-科学,かがく,science
-鏡,かがみ,mirror
-掛ける,かける,to hang something
-飾る,かざる,to decorate
-火事,かじ,fire
-,ガス,gas
-堅/硬/固い,かたい,hard
-形,かたち,shape
-片付ける,かたづける,to tidy up
-課長,かちょう,section manager
-勝つ,かつ,to win
-家内,かない,housewife
-悲しい,かなしい,sad
-必ず,かならず,"certainly,necessarily"
-お・金持ち,かねもち/おかねもち,rich man
-彼女,かのじょ,"she,girlfriend"
-壁,かべ,wall
-髪,かみ,hair
-噛む,かむ,"to bite,to chew"
-通う,かよう,to commute
-彼,かれ,"he,boyfriend"
-乾く,かわく,to get dry
-代わり,かわり,"substitute,alternate"
-変わる,かわる,to change
-考える,かんがえる,to consider
-関係,かんけい,relationship
-看護師,かんごし, nurse
-簡単,かんたん,simple
-気,き,"spirit,mood"
-機会,きかい,opportunity
-危険,きけん,danger
-聞こえる,きこえる,to be heard
-汽車,きしゃ,steam train
-技術,ぎじゅつ,"art,technology,skill"
-季節,きせつ,season
-規則,きそく,regulations
-,きっと,surely
-絹,きぬ,silk
-厳しい,きびしい,strict
-気分,きぶん,mood
-決る,きまる,to be decided
-君,きみ,(informal) You
-決める,きめる,to decide
-気持ち,きもち,"feeling,mood"
-着物,きもの,kimono
-客,きゃく,"guest,customer"
-急,きゅう,"urgent, steep"
-急行,きゅうこう,"speedy, express"
-教育,きょういく,education
-教会,きょうかい,church
-競争,きょうそう,competition
-興味,きょうみ,an interest
-近所,きんじょ,neighbourhood
-具合,ぐあい,"condition,health"
-空気,くうき,"air,atmosphere"
-空港,くうこう,airport
-草,くさ,grass
-首,くび,neck
-雲,くも,cloud
-比べる,くらべる,to compare
-,くれる,to give
-暮れる,くれる,"to get dark,to come to an end"
-君,くん,suffix for familiar young male
-毛,け,hair or fur
-経済,けいざい,"finance,economy"
-警察,けいさつ,police
-景色,けしき,"scene,landscape"
-消しゴム,けしゴム,eraser
-下宿,げしゅく,lodging
-決して,けっして,never
-,けれど/けれども,however
-原因,げんいん,"cause,source"
-,けんか・する,to quarrel
-研究,けんきゅう,research
-研究室,けんきゅうしつ,"study room,laboratory"
-見物,けんぶつ,sightseeing
-子,こ,child
-,こう,this way
-郊外,こうがい,outskirts
-講義,こうぎ,lecture
-工業,こうぎょう,the manufacturing industry
-高校,こうこう,high school
-高校生,こうこうせい,high school student
-工場,こうじょう/こうば,"factory,plant,mill,workshop"
-校長,こうちょう,headmaster
-交通,こうつう,"traffic,transportation"
-講堂,こうどう,auditorium
-高等学校,こうとうがっこう,high school
-公務員,こうむいん,"civil servant, government worker"
-国際,こくさい,international
-心,こころ,"heart, mind, core"
-御主人,ごしゅじん,(honorable) your husband
-故障,こしょう・する,to break-down
-ご存じ,ごぞんじ,(respect form ) to know
-答,こたえ,response
-,ごちそう,a feast
-小鳥,ことり,small bird
-,このあいだ,"the other day,recently"
-,このごろ,"these days,nowadays"
-細かい,こまかい,"small, fine"
-込む,こむ,to include
-米,こめ,uncooked rice
-,ごらんになる,(respectful) to see
-,これから,after this
-怖い,こわい,frightening
-壊す,こわす,to break
-壊れる,こわれる,to be broken
-今度,こんど,"now,next time"
-今夜,こんや,tonight
-最近,さいきん,"latest,nowadays"
-最後,さいご,"last,end"
-最初,さいしょ,"beginning,first"
-坂,さか,"slope,hill"
-探す,さがす,to look for
-下る,さがる,"to get down,to descend"
-盛ん,さかん,"popularity,prosperous"
-下げる,さげる,"to hang,to lower,to move back"
-差し上げる,さしあげる,(polite) to give
-,さっき,some time ago
-寂しい,さびしい,lonely
-さ来月,さらいげつ,the month after next
-さ来週,さらいしゅう,the week after next
-騒ぐ,さわぐ,"to make noise,to be excited"
-触る,さわる,to touch
-産業,さんぎょう,industry
-残念,ざんねん,disappointment
-市,し,city
-字,じ,character
-試合,しあい,"match,game"
-仕方,しかた,method
-試験,しけん,examination
-事故,じこ,accident
-地震,じしん,earthquake
-時代,じだい,era
-下着,したぎ,underwear
-,しっかり,"firmly,steadily"
-失敗,しっぱい,"failure,mistake"
-辞典,じてん,dictionary
-品物,しなもの,goods
-,しばらく,little while
-島,しま,island
-市民,しみん,citizen
-事務所,じむしょ,office
-社会,しゃかい,"society,public"
-社長,しゃちょう,company president
-自由,じゆう,freedom
-習慣,しゅうかん,"custom,manners"
-住所,じゅうしょ,"an address,a residence"
-柔道,じゅうどう,judo
-十分,じゅうぶん,enough
-趣味,しゅみ,hobby
-紹介,しょうかい,introduction
-小学校,しょうがっこう,elementary school
-小説,しょうせつ,novel
-将来,しょうらい,"future,prospects"
-食料品,しょくりょうひん,groceries
-女性,じょせい,woman
-知らせる,しらせる,to notify
-調べる,しらべる,to investigate
-人口,じんこう,population
-神社,じんじゃ,Shinto shrine
-親切,しんせつ,kindness
-新聞社,しんぶんしゃ,newspaper company
-水泳,すいえい,swimming
-水道,すいどう,water supply
-数学,すうがく,"mathematics,arithmetic"
-過ぎる,すぎる,to exceed
-凄い,すごい,terrific
-進む,すすむ,to make progress
-,すっかり,completely
-,すっと,"straight,all of a sudden"
-捨てる,すてる,to throw away
-砂,すな,sand
-滑る,すべる,"to slide,to slip"
-隅,すみ,"corner,nook"
-済む,すむ,to finish
-,すり,pickpocket
-,すると,then
-生活,せいかつ・する,to live
-生産,せいさん・する,to produce
-政治,せいじ,"politics,government"
-西洋,せいよう,western countries
-世界,せかい,the world
-席,せき,seat
-説明,せつめい,explanation
-背中,せなか,back of the body
-線,せん,line
-戦争,せんそう,war
-先輩,せんぱい,senior
-,そう,really
-育てる,そだてる,"to rear,to bring up"
-卒業,そつぎょう,graduation
-祖父,そふ,grandfather
-祖母,そぼ,grandmother
-,それで,because of that
-,それに,moreover
-,それほど,to that extent
-,そろそろ,"gradually,soon"
-,そんな,that sort of
-,そんなに,"so much,like that"
-退院,たいいん・する,to leave hospital
-大学生,だいがくせい,university student
-大事,だいじ,"important,valuable,serious matter"
-大体,だいたい,generally
-,たいてい,usually
-大分,だいぶ,greatly
-台風,たいふう,typhoon
-倒れる,たおれる,to break down
-,だから,"so,therefore"
-確か,たしか,definite
-足す,たす,to add a number
-訪ねる,たずねる,to visit
-尋ねる,たずねる,to ask
-正しい,ただしい,correct
-畳,たたみ,Japanese straw mat
-立てる,たてる,to stand something up
-建てる,たてる,to build
-例えば,たとえば,for example
-棚,たな,shelves
-楽しみ,たのしみ,joy
-楽む,たのしむ,to enjoy oneself
-,たまに,occasionally
-為,ため,in order to
-足りる,たりる,to be enough
-男性,だんせい,male
-暖房,だんぼう,heating
-血,ち,blood
-,チェック・する,to check
-力,ちから,"strength,power"
-,ちっとも,not at all (used with a negative verb)
-,ちゃん,suffix for familiar  person
-注意,ちゅうい,caution
-中学校,ちゅうがっこう,"junior high school,middle school"
-注射,ちゅうしゃ,injection
-駐車場,ちゅうしゃじょう,parking lot
-地理,ちり,geography
-捕まえる,つかまえる,to seize
-付く,つく,to be attached
-漬ける,つける,"to soak,to pickle"
-都合,つごう,"circumstances,convenience"
-伝える,つたえる,to report
-続く,つづく,to be continued
-続ける,つづける,to continue
-包む,つつむ,to wrap
-妻,つま,my wife
-,つもり,intention
-釣る,つる,to fish
-丁寧,ていねい,polite
-適当,てきとう,suitability
-手伝う,てつだう,to assist
-手袋,てぶくろ,glove
-寺,てら,temple
-点,てん,"point,dot"
-店員,てんいん,shop assistant
-天気予報,てんきよほう,weather forecast
-電灯,でんとう,electric light
-電報,でんぽう,telegram
-展覧会,てんらんかい,exhibition
-都,と,metropolitan
-道具,どうぐ,"tool,means"
-,とうとう,"finally, after all"
-動物園,どうぶつえん,zoo
-遠く,とおく,distant
-通る,とおる,to go through
-特に,とくに,"particularly,especially"
-特別,とくべつ,special
-,とこや,barber
-途中,とちゅう,on the way
-特急,とっきゅう,limited express train (faster than an express train)
-届ける,とどける,"to send, to deliver, to report"
-泊まる,とまる,to lodge at
-止める,とめる,to stop something
-取り替える,とりかえる,to exchange
-泥棒,どろぼう,thief
-,どんどん,more and more
-直す,なおす,"to fix,to repair"
-直る,なおる,"to be fixed,to be repaired"
-治る,なおる,"to be cured,to heal"
-泣く,なく,to weep
-無くなる,なくなる,"to disappear,to get lost"
-亡くなる,なくなる,to die
-投げる,なげる,to throw or cast away
-,なさる,(respectful) to do
-鳴る,なる,to sound
-,なるべく,as much as possible
-,なるほど,now I understand
-慣れる,なれる,to grow accustomed to
-苦い,にがい,bitter
-二階建て,にかいだて,two storied
-逃げる,にげる,to escape
-日記,にっき,journal
-入院,にゅういん・する,"to hospitalise, hospitalisation"
-入学,にゅうがく・する,to enter school or university
-似る,にる,to be similar
-人形,にんぎょう,"doll, figure"
-盗む,ぬすむ,to steal
-塗る,ぬる,"to paint, to colour, to plaster"
-,ぬれる,to get wet
-,ねだん,price
-熱,ねつ,fever
-寝坊,ねぼう,sleeping in late
-眠い,ねむい,sleepy
-眠る,ねむる,to sleep
-残る,のこる,to remain
-乗り換える,のりかえる,to change between buses or trains
-乗り物,のりもの,vehicle
-葉,は,leaf
-場合,ばあい,situation
-倍,ばい,double
-拝見,はいけん・する,(humble) to look at
-歯医者,はいしゃ,dentist
-運ぶ,はこぶ,to transport
-始める,はじめる,to begin
-場所,ばしょ,location
-,はず,it should be so
-恥ずかしい,はずかしい,embarrassed
-発音,はつおん,pronunciation
-,はっきり,clearly
-花見,はなみ,cherry-blossom viewing
-林,はやし,"woods,forester"
-払う,はらう,to pay
-番組,ばんぐみ,television or radio program
-反対,はんたい,opposition
-日,ひ,"day, sun"
-火,ひ,fire
-冷える,ひえる,to grow cold
-光,ひかり,light
-光る,ひかる,"to shine,to glitter"
-引き出し,ひきだし,"drawer,drawing out"
-,ひきだす,to withdraw
-,ひげ,beard
-飛行場,ひこうじょう,airport
-久しぶり,ひさしぶり,after a long time
-美術館,びじゅつかん,art gallery
-非常に,ひじょうに,extremely
-引っ越す,ひっこす,to move house
-必要,ひつよう,necessary
-,ひどい,awful
-開く,ひらく,to open an event
-昼間,ひるま,"daytime,during the day"
-昼休み,ひるやすみ,noon break
-拾う,ひろう,"to pick up,to gather"
-増える,ふえる,to increase
-深い,ふかい,deep
-複雑,ふくざつ,"complexity,complication"
-復習,ふくしゅう,revision
-部長,ぶちょう,head of a section
-普通,ふつう,"usually, or a train that stops at every station"
-,ぶどう,grapes
-太る,ふとる,to become fat
-布団,ふとん,"Japanese bedding, futon"
-舟,ふね,ship
-不便,ふべん,inconvenience
-踏む,ふむ,to step on
-降り出す,ふりだす,to start to rain
-文化,ぶんか,culture
-文学,ぶんがく,literature
-文法,ぶんぽう,grammar
-別,べつ,different
-変,へん,strange
-返事,へんじ,reply
-貿易,ぼうえき,trade
-法律,ほうりつ,law
-僕,ぼく,I (used by males)
-星,ほし,star
-,ほとんど,mostly
-,ほめる,to praise
-翻訳,ほんやく,translation
-参る,まいる,"(humble) to go,to come"
-負ける,まける,to lose
-,または,"or,otherwise"
-間違える,まちがえる,to make a mistake
-間に合う,まにあう,to be in time for
-周り,まわり,surroundings
-回る,まわる,to go around
-漫画,まんが,comic
-真中,まんなか,middle
-見える,みえる,to be in sight
-湖,みずうみ,lake
-味噌,みそ,"miso, soybean paste"
-見つかる,みつかる,to be discovered
-見つける,みつける,to discover
-皆,みな,everybody
-港,みなと,harbour
-向かう,むかう,to face
-迎える,むかえる,to go out to meet
-昔,むかし,"old times, old days, long ago, formerly"
-虫,むし,insect
-息子,むすこ,(humble) son
-娘,むすめ,(humble) daughter
-無理,むり,impossible
-召し上がる,めしあがる,(polite) to eat
-珍しい,めずらしい,rare
-申し上げる,もうしあげる,"(humble) to say,to tell"
-申す,もうす,"(humble) to be called,to say"
-,もうすぐ,soon
-,もし,if
-戻る,もどる,to turn back
-木綿,もめん,cotton
-森,もり,forest
-焼く,やく,"to bake,to grill"
-約束,やくそく,promise
-役に立つ,やくにたつ,to be helpful
-焼ける,やける,"to burn,to be roasted"
-優しい,やさしい,kind
-痩せる,やせる,to become thin
-,やっと,at last
-止む,やむ,to stop
-止める,やめる,to stop
-柔らかい,やわらかい,soft
-湯,ゆ,hot water
-指,ゆび,finger
-指輪,ゆびわ,a ring
-夢,ゆめ,dream
-揺れる,ゆれる,"to shake,to sway"
-用,よう,use
-用意,ようい,preparation
-用事,ようじ,things to do
-汚れる,よごれる,to get dirty
-予習,よしゅう,preparation for a lesson
-予定,よてい,arrangement
-予約,よやく,reservation
-寄る,よる,to visit
-喜ぶ,よろこぶ,to be delighted
-理由,りゆう,reason
-利用,りよう,utilization
-両方,りょうほう,both sides
-旅館,りょかん,Japanese hotel
-留守,るす,absence
-冷房,れいぼう,air conditioning
-歴史,れきし,history
-連絡,れんらく,contact
-沸かす,わかす,"to boil,to heat"
-別れる,わかれる,to separate
-沸く,わく,"to boil, to grow hot,to get excited"
-訳,わけ,"meaning,reason"
-忘れ物,わすれもの,lost article
-笑う,わらう,"to laugh,to smile"
-割合,わりあい,"rate,ratio,percentage"
-割れる,われる,to break
-,アクセサリー,accessory
-,アジア,Asia
-,アナウンサー,announcer
-,アフリカ,Africa
-,アメリカ,America
-,アルコール,alcohol
-,アルバイト,part-time job
-,エスカレーター,escalator
-,オートバイ,motorcycle
-,カーテン,curtain
-,ガス,gas
-,ガソリン,petrol
-,ガソリンスタンド,petrol station
-,ガラス,a glass pane
-,ケーキ,cake
-消しゴム,けしゴム,"eraser, rubber"
-,コンサート,concert
-,コンピューター,computer
-,サラダ,salad
-,サンダル,sandal
-,サンドイッチ,sandwich
-,ジャム,jam
-,スーツ,suit
-,スーツケース,suitcase
-,スクリーン,screen
-,ステーキ,steak
-,ステレオ,stereo
-,ソフト,soft
-,タイプ,"type,style"
-,チェック・する,to check
-,テキスト,"text,text book"
-,テニス,tennis
-,パート,part time
-,パソコン,personal computer
-,ハンドバッグ,handbag　
-,ピアノ,piano
-,ビル,building or bill
-,ファックス,fax
-,プレゼント,present
-,ベル,bell
-,レジ,register
-,レポート/リポート,report
-,ワープロ,word processor
--- a/data/tanos-jlpt/n5.csv
+++ b/data/tanos-jlpt/n5.csv
@@ -1,669 +0,0 @@
-会う,あう,to meet
-青,あお,blue
-青い,あおい,blue
-赤,あか,red
-赤い,あかい,red
-明い,あかるい,bright
-秋,あき,autumn
-開く,あく,"to open,to become open"
-開ける,あける,to open
-上げる,あげる,to give
-朝,あさ,morning
-朝御飯,あさごはん,breakfast
-,あさって,day after tomorrow
-足,あし,"foot,leg"
-明日,あした,tomorrow
-,あそこ,over there
-遊ぶ,あそぶ,"to play,to make a visit"
-暖かい,あたたかい,warm
-頭,あたま,head
-新しい,あたらしい,new
-,あちら,there
-暑い,あつい,hot
-熱い,あつい,hot to the touch
-厚い,あつい,"kind, deep, thick"
-,あっち,over there
-後,あと,afterwards
-,あなた,you
-兄,あに,(humble) older brother
-姉,あね,(humble) older sister
-,あの,that over there
-,あの,um...
-,アパート,apartment
-,あびる,"to bathe,to shower"
-危ない,あぶない,dangerous
-甘い,あまい,sweet
-,あまり,not very
-雨,あめ,rain
-飴,あめ,candy
-洗う,あらう,to wash
-,ある,"to be,to have (used for inanimate objects)"
-歩く,あるく,to walk
-,あれ,that
-,いい/よい,good
-,いいえ,no
-言う,いう,to say
-家,いえ,house
-,いかが,how
-行く,いく,to go
-,いくつ,"how many?,how old?"
-,いくら,how much?
-池,いけ,pond
-医者,いしゃ,medical doctor
-,いす,chair
-忙しい,いそがしい,"busy,irritated"
-痛い,いたい,painful
-一,いち,one
-一日,いちにち,"(1) one day, (2) first of month"
-,いちばん,"best,first"
-,いつ,when
-五日,いつか,"five days, fifth day"
-一緒,いっしょ,together
-五つ,いつつ,five
-,いつも,always
-犬,いぬ,dog
-今,いま,now
-意味,いみ,meaning
-妹,いもうと,(humble) younger sister
-嫌,いや,unpleasant
-入口,いりぐち,entrance
-居る,いる,"to be, to have (used for people and animals)"
-要る,いる,to need
-入れる,いれる,to put in
-色,いろ,colour
-,いろいろ,various
-上,うえ,on top of
-後ろ,うしろ,behind
-薄い,うすい,"thin,weak"
-歌,うた,song
-歌う,うたう,to sing
-生まれる,うまれる,to be born
-海,うみ,sea
-売る,うる,to sell
-煩い,うるさい,"noisy,annoying"
-上着,うわぎ,jacket
-絵,え,picture
-映画,えいが,movie
-映画館,えいがかん,cinema
-英語,えいご,English language
-,ええ,yes
-駅,えき,station
-,エレベーター,elevator
-鉛筆,えんぴつ,pencil
-,おいしい,delicious
-多い,おおい,many
-大きい,おおきい,big
-大きな,おおきな,big
-大勢,おおぜい,great number of people
-お母さん,おかあさん,(honorable) mother
-お菓子,おかし,"sweets, candy"
-お金,おかね,money
-起きる,おきる,to get up
-置く,おく,to put
-奥さん,おくさん,(honorable) wife
-お酒,おさけ,"alcohol, rice wine"
-お皿,おさら,"plate, dish"
-伯父/叔父,おじいさん,"grandfather,male senior citizen"
-教える,おしえる,"to teach,to tell"
-伯父/叔父,おじさん,"uncle,middle aged gentleman"
-押す,おす,"to push, to stamp something"
-遅い,おそい,"late,slow"
-お茶,おちゃ,green tea
-お手洗い,おてあらい,bathroom
-お父さん,おとうさん,(honorable) father
-弟,おとうと,younger brother
-男,おとこ,man
-男の子,おとこのこ,boy
-一昨日,おととい,day before yesterday
-一昨年,おととし,year before last
-大人,おとな,adult
-,おなか,stomach
-同じ,おなじ,same
-お兄さん,おにいさん,(honorable) older brother
-お姉さん,おねえさん,(honorable) older sister
-,おばあさん,"grandmother,female senior-citizen"
-伯母さん/叔母さん,おばさん,aunt
-お風呂,おふろ,bath
-お弁当,おべんとう,boxed lunch
-覚える,おぼえる,to remember
-,おまわりさん,friendly term for policeman
-重い,おもい,heavy
-,おもしろい,interesting
-泳ぐ,およぐ,to swim
-降りる,おりる,"to get off, to descend"
-終る,おわる,to finish
-音楽,おんがく,music
-女,おんな,woman
-女の子,おんなのこ,girl
-外国,がいこく,foreign country
-外国人,がいこくじん,foreigner
-会社,かいしゃ,company
-階段,かいだん,stairs
-買い物,かいもの,shopping
-買う,かう,to buy
-返す,かえす,to return something
-帰る,かえる,to go back
-,かかる,to take time or money
-,かぎ,key
-書く,かく,to write
-学生,がくせい,student
-,かける,to call by phone
-傘,かさ,umbrella
-貸す,かす,to lend
-風,かぜ,wind
-風邪,かぜ,a cold
-家族,かぞく,family
-方,かた,"person, way of doing"
-学校,がっこう,school
-,カップ,cup
-家庭,かてい,household
-角,かど,a corner
-,かばん,"bag,basket"
-花瓶,かびん,a vase
-紙,かみ,paper
-,カメラ,camera
-火曜日,かようび,Tuesday
-辛い,からい,spicy
-体,からだ,body
-借りる,かりる,to borrow
-軽い,かるい,light
-,カレー,curry
-,カレンダー,calendar
-川/河,かわ,river
-,かわいい,cute
-漢字,かんじ,Chinese character
-木,き,"tree,wood"
-黄色,きいろ,yellow
-黄色い,きいろい,yellow
-消える,きえる,to disappear
-聞く,きく,"to hear,to listen to,to ask"
-北,きた,north
-,ギター,guitar
-汚い,きたない,dirty
-喫茶店,きっさてん,coffee lounge
-切手,きって,postage stamp
-切符,きっぷ,ticket
-昨日,きのう,yesterday
-九,きゅう / く,nine
-牛肉,ぎゅうにく,beef
-牛乳,ぎゅうにゅう,milk
-今日,きょう,today
-教室,きょうしつ,classroom
-兄弟,きょうだい,(humble) siblings
-去年,きょねん,last year
-嫌い,きらい,hate
-切る,きる,to cut
-着る,きる,to put on from the shoulders down
-,きれい,"pretty,clean"
-,キロ/キログラム,kilogram
-,キロ/キロメートル,kilometre
-銀行,ぎんこう,bank
-金曜日,きんようび,Friday
-薬,くすり,medicine
-,ください,please
-果物,くだもの,fruit
-口,くち,"mouth,opening"
-靴,くつ,shoes
-靴下,くつした,socks
-国,くに,country
-曇り,くもり,cloudy weather
-曇る,くもる,"to become cloudy,to become dim"
-暗い,くらい,gloomy
-,クラス,class
-,グラム,gram
-来る,くる,to come
-車,くるま,"car,vehicle"
-黒,くろ,black
-黒い,くろい,black
-警官,けいかん,policeman
-今朝,けさ,this morning
-消す,けす,"to erase,to turn off power"
-結構,けっこう,"splendid,enough"
-結婚,けっこん,marriage
-月曜日,げつようび,Monday
-玄関,げんかん,entry hall
-元気,げんき,"health, vitality"
-五,ご,five
-公園,こうえん,park
-交差点,こうさてん,intersection
-紅茶,こうちゃ,black tea
-交番,こうばん,police box
-声,こえ,voice
-,コート,"coat,tennis court"
-,コーヒー,coffee
-,ここ,here
-午後,ごご,afternoon
-九日,ここのか,"nine days, ninth day"
-九つ,ここのつ,nine
-午前,ごぜん,morning
-答える,こたえる,to answer
-,こちら,this person or way
-,こっち,this person or way
-,コップ,a glass
-今年,ことし,this year
-言葉,ことば,"word,language"
-子供,こども,child
-,この,this
-御飯,ごはん,"cooked rice,meal"
-,コピーする,to copy
-困る,こまる,to be worried
-,これ,this
-今月,こんげつ,this month
-今週,こんしゅう,this week
-,こんな,such
-今晩,こんばん,this evening
-,さあ,well…
-財布,さいふ,wallet
-魚,さかな,fish
-先,さき,"the future,previous"
-咲く,さく,to bloom
-作文,さくぶん,"composition,writing"
-差す,さす,"to stretch out hands,to raise an umbrella"
-雑誌,ざっし,magazine
-砂糖,さとう,sugar
-寒い,さむい,cold
-さ来年,さらいねん,year after next
-三,さん,three
-散歩,さんぽする,to stroll
-四,し / よん,four
-塩,しお,salt
-,しかし,however
-時間,じかん,time
-仕事,しごと,job
-辞書,じしょ,dictionary
-静か,しずか,quiet
-下,した,below
-七,しち / なな,seven
-質問,しつもん,question
-自転車,じてんしゃ,bicycle
-自動車,じどうしゃ,automobile
-死ぬ,しぬ,to die
-字引,じびき,dictionary
-自分,じぶん,oneself
-閉まる,しまる,"to close,to be closed"
-閉める,しめる,to close something
-締める,しめる,to tie
-,じゃ/じゃあ,well then…
-写真,しゃしん,photograph
-,シャツ,shirt
-,シャワー,shower
-十,じゅう  とお,ten
-授業,じゅぎょう,"lesson,class work"
-宿題,しゅくだい,homework
-上手,じょうず,skillful
-丈夫,じょうぶ,"strong,durable"
-,しょうゆ,soy sauce
-食堂,しょくどう,dining hall
-知る,しる,to know
-白,しろ,white
-白い,しろい,white
-新聞,しんぶん,newspaper
-水曜日,すいようび,Wednesday
-吸う,すう,"to smoke,to suck"
-,スカート,skirt
-好き,すき,likeable
-少ない,すくない,a few
-,すぐに,instantly
-少し,すこし,few
-涼しい,すずしい,refreshing
-,ストーブ,heater
-,スプーン,spoon
-,スポーツ,sport
-,ズボン,trousers
-住む,すむ,to live in
-,スリッパ,slippers
-,する,to do
-座る,すわる,to sit
-背,せ,"height,stature"
-生徒,せいと,pupil
-,セーター,"sweater,jumper"
-,せっけん,soap
-背広,せびろ,business suit
-狭い,せまい,narrow
-,ゼロ,zero
-千,せん,thousand
-先月,せんげつ,last month
-先週,せんしゅう,last week
-先生,せんせい,"teacher,doctor"
-洗濯,せんたく,washing
-全部,ぜんぶ,all
-掃除,そうじする,"to clean, to sweep"
-,そうして/そして,and
-,そこ,that place
-,そちら,over there
-,そっち,over there
-外,そと,outside
-,その,that
-,そば,"near,beside"
-空,そら,sky
-,それ,that
-,それから,after that
-,それでは,in that situation
-大学,だいがく,university
-大使館,たいしかん,embassy
-大丈夫,だいじょうぶ,all right
-大好き,だいすき,to be very likeable
-大切,たいせつ,important
-台所,だいどころ,kitchen
-,たいへん,very
-,たいへん,difficult situation
-高い,たかい,"tall, expensive"
-,たくさん,many
-,タクシー,taxi
-出す,だす,to put out
-立つ,たつ,to stand
-,たて,"length,height"
-建物,たてもの,building
-楽しい,たのしい,enjoyable
-頼む,たのむ,to ask
-,たばこ,"tobacco,cigarettes"
-,たぶん,probably
-食べ物,たべもの,food
-食べる,たべる,to eat
-卵,たまご,egg
-誰,だれ,who
-誰,だれか,somebody
-誕生日,たんじょうび,birthday
-,だんだん,gradually
-小さい,ちいさい,little
-小さな,ちいさな,little
-近い,ちかい,near
-違う,ちがう,to differ
-近く,ちかく,near
-地下鉄,ちかてつ,underground train
-地図,ちず,map
-茶色,ちゃいろ,brown
-,ちゃわん,rice bowl
-,ちょうど,exactly
-,ちょっと,somewhat
-一日,ついたち,first of month
-使う,つかう,to use
-疲れる,つかれる,to get tired
-次,つぎ,next
-着く,つく,to arrive at
-机,つくえ,desk
-作る,つくる,to make
-,つける,to turn on
-勤める,つとめる,to work for someone
-,つまらない,boring
-冷たい,つめたい,cold to the touch
-強い,つよい,powerful
-手,て,hand
-,テープ,tape
-,テーブル,table
-,テープレコーダー,tape recorder
-出かける,でかける,to go out
-手紙,てがみ,letter
-,できる,to be able to
-出口,でぐち,exit
-,テスト,test
-,では,with that...
-,デパート,department store
-,でも,but
-出る,でる,"to appear,to leave"
-,テレビ,television
-天気,てんき,weather
-電気,でんき,"electricity,electric light"
-電車,でんしゃ,electric train
-電話,でんわ,telephone
-戸,と,Japanese style door
-,ドア,Western style door
-,トイレ,toilet
-,どう,"how,in what way"
-,どうして,for what reason
-,どうぞ,please
-動物,どうぶつ,animal
-,どうも,thanks
-遠い,とおい,far
-十日,とおか,"ten days,the tenth day"
-時々,ときどき,sometimes
-時計,とけい,"watch,clock"
-,どこ,where
-所,ところ,place
-年,とし,year
-図書館,としょかん,library
-,どちら,which of two
-,どっち,which
-,とても,very
-,どなた,who
-隣,となり,next door to
-,どの,which
-飛ぶ,とぶ,"to fly,to hop"
-止まる,とまる,to come to a halt
-友達,ともだち,friend
-土曜日,どようび,Saturday
-鳥,とり,bird
-とり肉,とりにく,chicken meat
-取る,とる,to take something
-撮る,とる,to take a photo or record a film
-,どれ,which (of three or more)
-,ナイフ,knife
-中,なか,middle
-長い,ながい,long
-鳴く,なく,"animal noise. to chirp, roar or croak etc."
-無くす,なくす,to lose something
-,なぜ,why
-夏,なつ,summer
-夏休み,なつやすみ,summer holiday
-,など,et cetera
-七つ,ななつ,seven
-七日,なのか,"seven days,the seventh day"
-名前,なまえ,name
-習う,ならう,to learn
-並ぶ,ならぶ,"to line up,to stand in a line"
-並べる,ならべる,"to line up,to set up"
-,なる,to become
-何,なん/なに,what
-二,に,two
-賑やか,にぎやか,"bustling,busy"
-肉,にく,meat
-西,にし,west
-日曜日,にちようび,Sunday
-荷物,にもつ,luggage
-,ニュース,news
-庭,にわ,garden
-脱ぐ,ぬぐ,to take off clothes
-温い,ぬるい,luke warm
-,ネクタイ,"tie,necktie"
-猫,ねこ,cat
-寝る,ねる,"to go to bed,to sleep"
-,ノート,"notebook,exercise book"
-登る,のぼる,to climb
-飲み物,のみもの,a drink
-飲む,のむ,to drink
-乗る,のる,"to get on,to ride"
-歯,は,tooth
-,パーティー,party
-,はい,yes
-灰皿,はいざら,ashtray
-入る,はいる,"to enter,to contain"
-葉書,はがき,postcard
-,はく,"to wear,to put on trousers"
-箱,はこ,box
-橋,はし,bridge
-,はし,chopsticks
-始まる,はじまる,to begin
-初め/始め,はじめ,beginning
-初めて,はじめて,for the first time
-走る,はしる,to run
-,バス,bus
-,バター,butter
-二十歳,はたち,"20 years old,20th year"
-働く,はたらく,to work
-八,はち,eight
-二十日,はつか,"twenty days,twentieth"
-花,はな,flower
-鼻,はな,nose
-話,はなし,"talk,story"
-話す,はなす,to speak
-早い,はやい,early
-速い,はやい,quick
-春,はる,spring
-貼る,はる,to stick
-晴れ,はれ,clear weather
-晴れる,はれる,to be sunny
-半,はん,half
-晩,ばん,evening
-,パン,bread
-,ハンカチ,handkerchief
-番号,ばんごう,number
-晩御飯,ばんごはん,evening meal
-半分,はんぶん,half minute
-東,ひがし,east
-引く,ひく,to pull
-弾く,ひく,"to play an instrument with strings, including piano"
-低い,ひくい,"short,low"
-飛行機,ひこうき,aeroplane
-左,ひだり,left hand side
-人,ひと,person
-一つ,ひとつ,one
-一月,ひとつき,one month
-一人,ひとり,one person
-暇,ひま,free time
-百,ひゃく,hundred
-病院,びょういん,hospital
-病気,びょうき,illness
-昼,ひる,"noon, daytime"
-昼御飯,ひるごはん,midday meal
-広い,ひろい,"spacious,wide"
-,フィルム,roll of film
-封筒,ふうとう,envelope
-,プール,swimming pool
-,フォーク,fork
-吹く,ふく,to blow
-服,ふく,clothes
-二つ,ふたつ,two
-豚肉,ぶたにく,pork
-二人,ふたり,two people
-二日,ふつか,"two days, second day of the month"
-太い,ふとい,fat
-冬,ふゆ,winter
-降る,ふる,"to fall, e.g. rain or snow"
-古い,ふるい,old (not used for people)
-,ふろ,bath
-文章,ぶんしょう,"sentence,text"
-,ページ,page
-下手,へた,unskillful
-,ベッド,bed
-,ペット,pet
-部屋,へや,room
-辺,へん,area
-,ペン,pen
-勉強,べんきょうする,to study
-便利,べんり,"useful, convenient"
-帽子,ぼうし,hat
-,ボールペン,ball-point pen
-,ほか,"other, the rest"
-,ポケット,pocket
-欲しい,ほしい,want
-,ポスト,post
-細い,ほそい,thin
-,ボタン,button
-,ホテル,hotel
-本,ほん,book
-本棚,ほんだな,bookshelves
-,ほんとう,truth
-毎朝,まいあさ,every morning
-毎月,まいげつ/まいつき,every month
-毎週,まいしゅう,every week
-毎日,まいにち,every day
-毎年,まいねん/まいとし,every year
-毎晩,まいばん,every night
-前,まえ,before
-曲る,まがる,"to turn,to bend"
-,まずい,unpleasant
-,また,"again,and"
-,まだ,"yet,still"
-町,まち,"town,city"
-待つ,まつ,to wait
-,まっすぐ,"straight ahead,direct"
-,マッチ,match
-窓,まど,window
-丸い/円い,まるい,"round,circular"
-万,まん,ten thousand
-万年筆,まんねんひつ,fountain pen
-磨く,みがく,"to brush teeth, to polish"
-右,みぎ,right side
-短い,みじかい,short
-水,みず,water
-店,みせ,shop
-見せる,みせる,to show
-道,みち,street
-三日,みっか,"three days, third day of the month"
-三つ,みっつ,three
-緑,みどり,green
-皆さん,みなさん,everyone
-南,みなみ,south
-耳,みみ,ear
-見る  観る,みる,"to see, to watch"
-,みんな,everyone
-六日,むいか,"six days, sixth day of the month"
-向こう,むこう,over there
-難しい,むずかしい,difficult
-六つ,むっつ,six
-村,むら,village
-目,め,eye
-,メートル,metre
-眼鏡,めがね,glasses
-,もう,already
-もう一度,もういちど,again
-木曜日,もくようび,Thursday
-持つ,もつ,to hold
-,もっと,more
-物,もの,thing
-門,もん,gate
-問題,もんだい,problem
-八百屋,やおや,greengrocer
-野菜,やさい,vegetable
-易しい,やさしい,"easy, simple"
-安い,やすい,cheap
-休み,やすみ,"rest,holiday"
-休む,やすむ,to rest
-八つ,やっつ,eight
-山,やま,mountain
-,やる,to do
-夕方,ゆうがた,evening
-夕飯,ゆうはん,dinner
-郵便局,ゆうびんきょく,post office
-昨夜,ゆうべ,last night
-有名,ゆうめい,famous
-雪,ゆき,snow
-行く,ゆく,to go
-,ゆっくりと,slowly
-八日,ようか,"eight days, eighth day of the month"
-洋服,ようふく,western-style clothes
-,よく,"often, well"
-横,よこ,"beside,side,width"
-四日,よっか,"four days, fouth day of the month"
-四つ,よっつ,four
-呼ぶ,よぶ,"to call out,to invite"
-読む,よむ,to read
-夜,よる,"evening,night"
-弱い,よわい,weak
-来月,らいげつ,next month
-来週,らいしゅう,next week
-来年,らいねん,next year
-,ラジオ,radio
-,ラジカセ / ラジオカセット,radio cassette player
-,りっぱ,splendid
-留学生,りゅうがくせい,overseas student
-両親,りょうしん,both parents
-料理,りょうり,cuisine
-旅行,りょこう,travel
-零,れい,zero
-冷蔵庫,れいぞうこ,refrigerator
-,レコード,record
-,レストラン,restaurant
-練習,れんしゅうする,to practice
-廊下,ろうか,corridor
-六,ろく,six
-,ワイシャツ,business shirt
-若い,わかい,young
-分かる,わかる,to be understood
-忘れる,わすれる,to forget
-私,わたくし,"(humble) I,myself"
-私,わたし,"I,myself"
-渡す,わたす,to hand over
-渡る,わたる,to go across
-悪い,わるい,bad
-,より、ほう,Used for comparison.
--- a/docs/database.md
+++ b/docs/database.md
@@ -0,0 +1,29 @@
+# Database
+
+Here are some choices that have been made when designing the schema
+
+### `JMdict_{Reading,Kanji}Element.elementId` and `JMdict_Sense.senseId`
+
+The `elementId`/`senseId` field acts as a unique identifier for each individual element in these tables.
+It is a packed version of the `(entryId, orderNum)` pair, where the first number is given 7 digits and the second is given 2 digits (max count found so far is `40`).
+Since `entryId` already is a field in the table, it would technically have been fine to store the `orderNum` as a separate field,
+but it is easier to be able to refer to the entries without a composite foreign key in other tables.
+
+(NOTE: `entryId` is now inferred from `elementId` within sqlite using a generated column, so saying it is "stored in a separate field" might be a stretch)
+
+In addition, the reading element id's are added with `1000000000` to make them unique from the kanji element id's. This reduces the amount of space needed for indices in some locations, because you can simply filter out each part with `>` or `<`.
+
+We used to generate the `elementId` separately from `orderNum` as a sequential id, but it lead to all values
+shifting whenever the data was updated, leading to very big diffs. Making it be a unique composite of data coming
+from the source data itself means that the values will be stable across updates.
+
+Due to the way the data is structured, we can use the `elementId` as the ordering number as well.
+
+### `JMdict_EntryScore`
+
+The `JMdict_EntryScore` table is used to store the score of each entry, which is used for sorting search results. The score is calculated based on a number of variables.
+
+The table is automatically generated from other tables via triggers, and should be considered as a materialized view.
+
+<s>There is a score row for every single entry in both `JMdict_KanjiElement` and `JMdict_ReadingElement`, split by the `type` field.</s>
+This is no longer true, we now only store the rows for which the score is not `0`. The `type` field is now also virtual, since the `elementId` fields for both kanji and readings are unique to each other.
--- a/docs/lemmatizer.md
+++ b/docs/lemmatizer.md
@@ -0,0 +1,13 @@
+# Lemmatizer
+
+The lemmatizer is still quite experimental, but will play a more important role in the project in the future.
+
+It is a manual implementation of a [Finite State Transducer](https://en.wikipedia.org/wiki/Morphological_dictionary#Finite_State_Transducers) for morphological parsing. The FST is used to recursively remove affixes from a word until it (hopefully) deconjugates into its dictionary form. This iterative deconjugation tree will then be combined with queries into the dictionary data to determine if the deconjugation leads to a real known word.
+
+Each separate rule is a separate static object declared in `lib/util/lemmatizer/rules`.
+
+There is a cli subcommand for testing the tool interactively, you can run
+
+```bash
+dart run jadb lemmatize -w '食べさせられない'
+```
--- a/docs/overview.md
+++ b/docs/overview.md
@@ -0,0 +1,28 @@
+# Overview
+
+This is the documentation for `jadb`. Since I'm currently the only one working on it, the documentation is more or less just notes to myself, to ensure I remember how and why I implemented certain features in a certain way a few months down the road. This is not a comprehensive and formal documentation for downstream use, neither for developers nor end-users.
+
+- [Word Search](./word-search.md)
+- [Database](./database.md)
+- [Lemmatizer](./lemmatizer.md)
+
+## Project structure
+
+- `lib/_data_ingestion` contains all the code for reading data sources, transforming them and compiling them into an SQLite database. This is for the most part isolated from the rest of the codebase, and should not be depended on by any code used for querying the database.
+- `lib/cli` contains code for cli tooling (e.g. argument parsing, subcommand handling, etc.)
+- `lib/const_data` contains database data that is small enough to warrant being hardcoded as dart constants.
+- `lib/models` contains all the code for representing the database schema as Dart classes, and for converting between those classes and the actual database.
+- `lib/search` contains all the code for searching the database.
+- `lib/util/lemmatizer` contains the code for lemmatization, which will be used by the search code in the future.
+- `migrations` contains raw SQL files for creating the database schema.
+
+## SQLite naming conventions
+
+> [!WARNING]
+> All of these conventions are actually not enforced yet, it will be fixed at some point.
+
+- Indices are prefixed with `IDX__`
+- Crossref tables are prefixed with `XREF__`
+- Trigger names are prefixed with `TRG__`
+- Views are prefixed with `VW__`
+- All data sources should have a `<datasource>_Version` table, which contains a single row with the version of the data source used to generate the database.
--- a/docs/word-search.md
+++ b/docs/word-search.md
@@ -0,0 +1,21 @@
+# Word search
+
+The word search procedure is currently split into 3 parts:
+
+1. **Entry ID query**:
+
+Use a complex query with various scoring factors to try to get list of
+database ids pointing at dictionary entries, sorted by how likely we think this
+word is the word that the caller is looking for. The output here is a `List<int>`
+
+2. **Data Query**:
+
+Takes the entry id list from the last search, and performs all queries needed to retrieve
+all the dictionary data for those IDs. The result is a struct with a bunch of flattened lists
+with data for all the dictionary entries. These lists are sorted by the order that the ids
+were provided.
+
+3. **Regrouping**:
+
+Takes the flattened data, and regroups the items into structs with a more "hierarchical" structure.
+All data tagged with the same ID will end up in the same struct. Returns a list of these structs.
--- a/flake.lock
+++ b/flake.lock
@@ -1,48 +1,32 @@
 {
  "nodes": {
-    "jmdict-src": {
-      "flake": false,
+    "datasources": {
+      "inputs": {
+        "nixpkgs": [
+          "nixpkgs"
+        ]
+      },
      "locked": {
-        "narHash": "sha256-lh46uougUzBrRhhwa7cOb32j5Jt9/RjBUhlVjwVzsII=",
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz"
+        "lastModified": 1776081209,
+        "narHash": "sha256-zR1115tcOPnYLk6NznSf7YslyaJLc/MGayEHShitx18=",
+        "ref": "refs/heads/main",
+        "rev": "7fe3552bb16e1d315c0b27b243e5eb53cd9e86fc",
+        "revCount": 13,
+        "type": "git",
+        "url": "https://git.pvv.ntnu.no/Mugiten/datasources.git"
      },
      "original": {
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz"
-      }
-    },
-    "jmdict-with-examples-src": {
-      "flake": false,
-      "locked": {
-        "narHash": "sha256-5oS2xDyetbuSM6ax3LUjYA3N60x+D3Hg41HEXGFMqLQ=",
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e_examp.gz"
-      },
-      "original": {
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/JMdict_e_examp.gz"
-      }
-    },
-    "kanjidic2-src": {
-      "flake": false,
-      "locked": {
-        "narHash": "sha256-orSeQqSxhn9TtX3anYtbiMEm7nFkuomGnIKoVIUR2CM=",
-        "type": "file",
-        "url": "https://www.edrdg.org/kanjidic/kanjidic2.xml.gz"
-      },
-      "original": {
-        "type": "file",
-        "url": "https://www.edrdg.org/kanjidic/kanjidic2.xml.gz"
+        "type": "git",
+        "url": "https://git.pvv.ntnu.no/Mugiten/datasources.git"
      }
    },
    "nixpkgs": {
      "locked": {
-        "lastModified": 1771848320,
-        "narHash": "sha256-0MAd+0mun3K/Ns8JATeHT1sX28faLII5hVLq0L3BdZU=",
+        "lastModified": 1775423009,
+        "narHash": "sha256-vPKLpjhIVWdDrfiUM8atW6YkIggCEKdSAlJPzzhkQlw=",
        "owner": "NixOS",
        "repo": "nixpkgs",
-        "rev": "2fc6539b481e1d2569f25f8799236694180c0993",
+        "rev": "68d8aa3d661f0e6bd5862291b5bb263b2a6595c9",
        "type": "github"
      },
      "original": {
@@ -51,25 +35,10 @@
        "type": "indirect"
      }
    },
-    "radkfile-src": {
-      "flake": false,
-      "locked": {
-        "narHash": "sha256-DHpMUE2Umje8PbzXUCS6pHZeXQ5+WTxbjSkGU3erDHQ=",
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/radkfile.gz"
-      },
-      "original": {
-        "type": "file",
-        "url": "http://ftp.edrdg.org/pub/Nihongo/radkfile.gz"
-      }
-    },
    "root": {
      "inputs": {
-        "jmdict-src": "jmdict-src",
-        "jmdict-with-examples-src": "jmdict-with-examples-src",
-        "kanjidic2-src": "kanjidic2-src",
-        "nixpkgs": "nixpkgs",
-        "radkfile-src": "radkfile-src"
+        "datasources": "datasources",
+        "nixpkgs": "nixpkgs"
      }
    }
  },
--- a/flake.nix
+++ b/flake.nix
@@ -4,35 +4,16 @@
  inputs = {
    nixpkgs.url = "nixpkgs/nixos-unstable";

-    jmdict-src = {
-      # url = "http://ftp.edrdg.org/pub/Nihongo/JMdict.gz";
-      url = "http://ftp.edrdg.org/pub/Nihongo/JMdict_e.gz";
-      flake = false;
-    };
-
-    jmdict-with-examples-src = {
-      url = "http://ftp.edrdg.org/pub/Nihongo/JMdict_e_examp.gz";
-      flake = false;
-    };
-
-    radkfile-src = {
-      url = "http://ftp.edrdg.org/pub/Nihongo/radkfile.gz";
-      flake = false;
-    };
-
-    kanjidic2-src = {
-      url = "https://www.edrdg.org/kanjidic/kanjidic2.xml.gz";
-      flake = false;
+    datasources = {
+      url = "git+https://git.pvv.ntnu.no/Mugiten/datasources.git";
+      inputs.nixpkgs.follows = "nixpkgs";
    };
  };

  outputs = {
    self,
    nixpkgs,
-    jmdict-src,
-    jmdict-with-examples-src,
-    radkfile-src,
-    kanjidic2-src
+    datasources,
  }: let
    inherit (nixpkgs) lib;
    systems = [
@@ -77,15 +58,12 @@

    devShells = forAllSystems (system: pkgs: {
      default = pkgs.mkShell {
-        buildInputs = with pkgs; [
+        packages = with pkgs; [
          dart
          gnumake
          lcov
-          sqlite-analyzer
+          sqldiff
          sqlite-interactive
-          sqlite-web
-          # sqlint
-          sqlfluff
        ];
        env = {
          LIBSQLITE_PATH = "${pkgs.sqlite.out}/lib/libsqlite3.so";
@@ -93,6 +71,16 @@
          LD_LIBRARY_PATH = lib.makeLibraryPath [ pkgs.sqlite ];
        };
      };
+
+      sqlite-debugging = pkgs.mkShell {
+        packages = with pkgs; [
+          sqlite-interactive
+          sqlite-analyzer
+          sqlite-web
+          # sqlint
+          sqlfluff
+        ];
+      };
    });

    packages = let
@@ -128,29 +116,21 @@
        ln -s ${src} $out
      '';

-      jmdict = pkgs.callPackage ./nix/jmdict.nix {
-        inherit jmdict-src jmdict-with-examples-src edrdgMetadata;
-      };
-
-      radkfile =  pkgs.callPackage ./nix/radkfile.nix {
-        inherit radkfile-src edrdgMetadata;
-      };
-
-      kanjidic2 = pkgs.callPackage ./nix/kanjidic2.nix {
-        inherit kanjidic2-src edrdgMetadata;
-      };
+      inherit (datasources.packages.${system}) jmdict radkfile kanjidic2;

      database-tool = pkgs.callPackage ./nix/database_tool.nix {
        inherit src;
      };

      database = pkgs.callPackage ./nix/database.nix {
-        inherit (self.packages.${system}) database-tool jmdict radkfile kanjidic2;
+        inherit (datasources.packages.${system}) jmdict radkfile kanjidic2 tanos-jlpt;
+        inherit (self.packages.${system}) database-tool;
        inherit src;
      };

      database-wal = pkgs.callPackage ./nix/database.nix {
-        inherit (self.packages.${system}) database-tool jmdict radkfile kanjidic2;
+        inherit (datasources.packages.${system}) jmdict radkfile kanjidic2 tanos-jlpt;
+        inherit (self.packages.${system}) database-tool;
        inherit src;
        wal = true;
      };
--- a/lib/_data_ingestion/jmdict/objects.dart
+++ b/lib/_data_ingestion/jmdict/objects.dart
@@ -1,13 +1,15 @@
 import 'package:jadb/_data_ingestion/sql_writable.dart';

 abstract class Element extends SQLWritable {
+  final int elementId;
  final String reading;
  final int? news;
  final int? ichi;
  final int? spec;
  final int? gai;
  final int? nf;
-  const Element({
+  Element({
+    required this.elementId,
    required this.reading,
    this.news,
    this.ichi,
@@ -18,6 +20,7 @@ abstract class Element extends SQLWritable {

  @override
  Map<String, Object?> get sqlValue => {
+    'elementId': elementId,
    'reading': reading,
    'news': news,
    'ichi': ichi,
@@ -28,12 +31,11 @@ abstract class Element extends SQLWritable {
 }

 class KanjiElement extends Element {
-  int orderNum;
  List<String> info;

  KanjiElement({
    this.info = const [],
-    required this.orderNum,
+    required super.elementId,
    required super.reading,
    super.news,
    super.ichi,
@@ -45,21 +47,19 @@ class KanjiElement extends Element {
  @override
  Map<String, Object?> get sqlValue => {
    ...super.sqlValue,
-    'orderNum': orderNum,
  };
 }

 class ReadingElement extends Element {
-  int orderNum;
  bool readingDoesNotMatchKanji;
  List<String> info;
  List<String> restrictions;

  ReadingElement({
-    required this.orderNum,
    required this.readingDoesNotMatchKanji,
    this.info = const [],
    this.restrictions = const [],
+    required super.elementId,
    required super.reading,
    super.news,
    super.ichi,
@@ -71,7 +71,6 @@ class ReadingElement extends Element {
  @override
  Map<String, Object?> get sqlValue => {
    ...super.sqlValue,
-    'orderNum': orderNum,
    'readingDoesNotMatchKanji': readingDoesNotMatchKanji,
  };
 }
@@ -107,9 +106,8 @@ class Glossary extends SQLWritable {

  @override
  Map<String, Object?> get sqlValue => {
-    'language': language,
+    // 'language': language,
    'phrase': phrase,
-    'type': type,
  };
 }

@@ -142,7 +140,6 @@ class XRef {

 class Sense extends SQLWritable {
  final int senseId;
-  final int orderNum;
  final List<XRefParts> antonyms;
  final List<String> dialects;
  final List<String> fields;
@@ -157,7 +154,6 @@ class Sense extends SQLWritable {

  const Sense({
    required this.senseId,
-    required this.orderNum,
    this.antonyms = const [],
    this.dialects = const [],
    this.fields = const [],
@@ -174,7 +170,6 @@ class Sense extends SQLWritable {
  @override
  Map<String, Object?> get sqlValue => {
    'senseId': senseId,
-    'orderNum': orderNum,
  };

  bool get isEmpty =>
--- a/lib/_data_ingestion/jmdict/seed_data.dart
+++ b/lib/_data_ingestion/jmdict/seed_data.dart
@@ -1,17 +1,27 @@
 import 'dart:collection';
+import 'dart:io';

 import 'package:collection/collection.dart';
 import 'package:jadb/_data_ingestion/jmdict/objects.dart';
 import 'package:jadb/table_names/jmdict.dart';
 import 'package:sqflite_common/sqlite_api.dart';

+/// A wrapper for the result of resolving an xref, which includes the resolved entry and a flag
+/// indicating whether the xref was ambiguous (i.e. could refer to multiple entries).
 class ResolvedXref {
-  Entry entry;
-  bool ambiguous;
+  final Entry entry;
+  final bool ambiguous;

-  ResolvedXref(this.entry, this.ambiguous);
+  const ResolvedXref(this.entry, this.ambiguous);
 }

+/// Resolves an xref (pair of kanji, optionally reading, and optionally sense number) to an a specific
+/// JMdict entry, if possible.
+///
+/// If the xref is ambiguous (i.e. it could refer to multiple entries), the
+/// first entry is returned, and the returned value is marked as ambiguous.
+///
+/// If the xref cannot be resolved to any entry at all, an exception is thrown.
 ResolvedXref resolveXref(
  SplayTreeMap<String, Set<Entry>> entriesByKanji,
  SplayTreeMap<String, Set<Entry>> entriesByReading,
@@ -75,41 +85,47 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {
  print('  [JMdict] Batch 1 - Kanji and readings');
  Batch b = db.batch();

-  int elementId = 0;
+  if (Platform.environment['JMDICT_VERSION'] != null &&
+      Platform.environment['JMDICT_DATE'] != null &&
+      Platform.environment['JMDICT_HASH'] != null) {
+    b.insert(JMdictTableNames.version, {
+      'version': Platform.environment['JMDICT_VERSION']!,
+      'date': Platform.environment['JMDICT_DATE']!,
+      'hash': Platform.environment['JMDICT_HASH']!,
+    });
+  } else {
+    print(
+      'WARNING: JMDICT version information not found in environment variables. '
+      'This may cause issues with future updates.',
+    );
+  }
+
  for (final e in entries) {
    b.insert(JMdictTableNames.entry, e.sqlValue);

    for (final k in e.kanji) {
-      elementId++;
-      b.insert(
-        JMdictTableNames.kanjiElement,
-        k.sqlValue..addAll({'entryId': e.entryId, 'elementId': elementId}),
-      );
+      b.insert(JMdictTableNames.kanjiElement, k.sqlValue);

      for (final i in k.info) {
        b.insert(JMdictTableNames.kanjiInfo, {
-          'elementId': elementId,
+          'elementId': k.elementId,
          'info': i,
        });
      }
    }

    for (final r in e.readings) {
-      elementId++;
-      b.insert(
-        JMdictTableNames.readingElement,
-        r.sqlValue..addAll({'entryId': e.entryId, 'elementId': elementId}),
-      );
+      b.insert(JMdictTableNames.readingElement, r.sqlValue);

      for (final i in r.info) {
        b.insert(JMdictTableNames.readingInfo, {
-          'elementId': elementId,
+          'elementId': r.elementId,
          'info': i,
        });
      }
      for (final res in r.restrictions) {
        b.insert(JMdictTableNames.readingRestriction, {
-          'elementId': elementId,
+          'elementId': r.elementId,
          'restriction': res,
        });
      }
@@ -123,10 +139,7 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {

  for (final e in entries) {
    for (final s in e.senses) {
-      b.insert(
-        JMdictTableNames.sense,
-        s.sqlValue..addAll({'entryId': e.entryId}),
-      );
+      b.insert(JMdictTableNames.sense, s.sqlValue);
      for (final d in s.dialects) {
        b.insert(JMdictTableNames.senseDialect, {
          'senseId': s.senseId,
@@ -150,16 +163,18 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {
      }
      for (final rk in s.restrictedToKanji) {
        b.insert(JMdictTableNames.senseRestrictedToKanji, {
-          'entryId': e.entryId,
          'senseId': s.senseId,
-          'kanji': rk,
+          'kanjiElementId': e.kanji
+              .firstWhere((k) => k.reading == rk)
+              .elementId,
        });
      }
      for (final rr in s.restrictedToReading) {
        b.insert(JMdictTableNames.senseRestrictedToReading, {
-          'entryId': e.entryId,
          'senseId': s.senseId,
-          'reading': rr,
+          'readingElementId': e.readings
+              .firstWhere((r) => r.reading == rr)
+              .elementId,
        });
      }
      for (final ls in s.languageSource) {
@@ -173,6 +188,14 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {
          JMdictTableNames.senseGlossary,
          g.sqlValue..addAll({'senseId': s.senseId}),
        );
+
+        if (g.type != null) {
+          b.insert(JMdictTableNames.senseGlossaryType, {
+            'senseId': s.senseId,
+            'phrase': g.phrase,
+            'type': g.type!,
+          });
+        }
      }
    }
  }
@@ -181,24 +204,17 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {

  print('  [JMdict] Building xref trees');
  final SplayTreeMap<String, Set<Entry>> entriesByKanji = SplayTreeMap();
+  final SplayTreeMap<String, Set<Entry>> entriesByReading = SplayTreeMap();

  for (final entry in entries) {
    for (final kanji in entry.kanji) {
-      if (entriesByKanji.containsKey(kanji.reading)) {
-        entriesByKanji.update(kanji.reading, (list) => list..add(entry));
-      } else {
-        entriesByKanji.putIfAbsent(kanji.reading, () => {entry});
-      }
+      entriesByKanji.putIfAbsent(kanji.reading, () => {});
+      entriesByKanji.update(kanji.reading, (set) => set..add(entry));
    }
-  }
-  final SplayTreeMap<String, Set<Entry>> entriesByReading = SplayTreeMap();
-  for (final entry in entries) {
+
    for (final reading in entry.readings) {
-      if (entriesByReading.containsKey(reading.reading)) {
-        entriesByReading.update(reading.reading, (list) => list..add(entry));
-      } else {
-        entriesByReading.putIfAbsent(reading.reading, () => {entry});
-      }
+      entriesByReading.putIfAbsent(reading.reading, () => {});
+      entriesByReading.update(reading.reading, (set) => set..add(entry));
    }
  }

@@ -207,6 +223,7 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {

  for (final e in entries) {
    for (final s in e.senses) {
+      final seenSeeAlsoXrefs = <int>{};
      for (final xref in s.seeAlso) {
        final resolvedEntry = resolveXref(
          entriesByKanji,
@@ -214,16 +231,26 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {
          xref,
        );

+        if (seenSeeAlsoXrefs.contains(resolvedEntry.entry.entryId)) {
+          print(
+            'WARNING: Skipping duplicate seeAlso xref from sense ${s.senseId} to entry ${resolvedEntry.entry.entryId}\n'
+            '  (kanjiRef: ${xref.kanjiRef}, readingRef: ${xref.readingRef}, senseOrderNum: ${xref.senseOrderNum})',
+          );
+          continue;
+        }
+        seenSeeAlsoXrefs.add(resolvedEntry.entry.entryId);
+
        b.insert(JMdictTableNames.senseSeeAlso, {
          'senseId': s.senseId,
          'xrefEntryId': resolvedEntry.entry.entryId,
-          'seeAlsoKanji': xref.kanjiRef,
-          'seeAlsoReading': xref.readingRef,
-          'seeAlsoSense': xref.senseOrderNum,
+          'seeAlsoSense': xref.senseOrderNum != null
+              ? xref.senseOrderNum! - 1
+              : null,
          'ambiguous': resolvedEntry.ambiguous,
        });
      }

+      final seenAntonymXrefs = <int>{};
      for (final ant in s.antonyms) {
        final resolvedEntry = resolveXref(
          entriesByKanji,
@@ -231,12 +258,21 @@ Future<void> seedJMDictData(List<Entry> entries, Database db) async {
          ant,
        );

+        if (seenAntonymXrefs.contains(resolvedEntry.entry.entryId)) {
+          print(
+            'WARNING: Skipping duplicate antonym xref from sense ${s.senseId} to entry ${resolvedEntry.entry.entryId}\n'
+            '  (kanjiRef: ${ant.kanjiRef}, readingRef: ${ant.readingRef}, senseOrderNum: ${ant.senseOrderNum})',
+          );
+          continue;
+        }
+        seenAntonymXrefs.add(resolvedEntry.entry.entryId);
+
        b.insert(JMdictTableNames.senseAntonyms, {
          'senseId': s.senseId,
          'xrefEntryId': resolvedEntry.entry.entryId,
-          'antonymKanji': ant.kanjiRef,
-          'antonymReading': ant.readingRef,
-          'antonymSense': ant.senseOrderNum,
+          'antonymSense': ant.senseOrderNum != null
+              ? ant.senseOrderNum! - 1
+              : null,
          'ambiguous': resolvedEntry.ambiguous,
        });
      }
--- a/lib/_data_ingestion/jmdict/xml_parser.dart
+++ b/lib/_data_ingestion/jmdict/xml_parser.dart
@@ -71,8 +71,6 @@ XRefParts parseXrefParts(String s) {
 List<Entry> parseJMDictData(XmlElement root) {
  final List<Entry> entries = [];

-  int senseId = 0;
-
  for (final entry in root.childElements) {
    final entryId = int.parse(entry.findElements('ent_seq').first.innerText);

@@ -80,11 +78,18 @@ List<Entry> parseJMDictData(XmlElement root) {
    final List<ReadingElement> readingEls = [];
    final List<Sense> senses = [];

-    for (final (kanjiNum, kEle) in entry.findElements('k_ele').indexed) {
+    for (final (orderNum, kEle) in entry.findElements('k_ele').indexed) {
+      assert(
+        orderNum < 100,
+        'Entry $entryId has more than 100 kanji elements, which will break the elementId generation logic.',
+      );
+      final elementId = entryId * 100 + orderNum;
+
      final kePri = getPriorityValues(kEle, 'ke');
+
      kanjiEls.add(
        KanjiElement(
-          orderNum: kanjiNum + 1,
+          elementId: elementId,
          info: kEle
              .findElements('ke_inf')
              .map((e) => e.innerText.substring(1, e.innerText.length - 1))
@@ -100,13 +105,20 @@ List<Entry> parseJMDictData(XmlElement root) {
    }

    for (final (orderNum, rEle) in entry.findElements('r_ele').indexed) {
+      assert(
+        orderNum < 100,
+        'Entry $entryId has more than 100 readings, which will break the elementId generation logic.',
+      );
+      final elementId = 1_000_000_000 + entryId * 100 + orderNum;
+
      final rePri = getPriorityValues(rEle, 're');
      final readingDoesNotMatchKanji = rEle
          .findElements('re_nokanji')
          .isNotEmpty;
+
      readingEls.add(
        ReadingElement(
-          orderNum: orderNum + 1,
+          elementId: elementId,
          readingDoesNotMatchKanji: readingDoesNotMatchKanji,
          info: rEle
              .findElements('re_inf')
@@ -127,10 +139,14 @@ List<Entry> parseJMDictData(XmlElement root) {
    }

    for (final (orderNum, sense) in entry.findElements('sense').indexed) {
-      senseId++;
+      assert(
+        orderNum < 100,
+        'Entry $entryId has more than 100 senses, which will break the senseId generation logic.',
+      );
+      final senseId = entryId * 100 + orderNum;
+
      final result = Sense(
        senseId: senseId,
-        orderNum: orderNum + 1,
        restrictedToKanji: sense
            .findElements('stagk')
            .map((e) => e.innerText)
--- a/lib/_data_ingestion/kanjidic/objects.dart
+++ b/lib/_data_ingestion/kanjidic/objects.dart
@@ -248,9 +248,6 @@ class Character extends SQLWritable {
  @override
  Map<String, Object?> get sqlValue => {
    'literal': literal,
-    'grade': grade,
    'strokeCount': strokeCount,
-    'frequency': frequency,
-    'jlpt': jlpt,
  };
 }
--- a/lib/_data_ingestion/kanjidic/seed_data.dart
+++ b/lib/_data_ingestion/kanjidic/seed_data.dart
@@ -1,3 +1,5 @@
+import 'dart:io';
+
 import 'package:jadb/table_names/kanjidic.dart';
 import 'package:sqflite_common/sqlite_api.dart';

@@ -5,6 +7,22 @@ import 'objects.dart';

 Future<void> seedKANJIDICData(List<Character> characters, Database db) async {
  final b = db.batch();
+
+  if (Platform.environment['KANJIDIC_VERSION'] != null &&
+      Platform.environment['KANJIDIC_DATE'] != null &&
+      Platform.environment['KANJIDIC_HASH'] != null) {
+    b.insert(KANJIDICTableNames.version, {
+      'version': Platform.environment['KANJIDIC_VERSION']!,
+      'date': Platform.environment['KANJIDIC_DATE']!,
+      'hash': Platform.environment['KANJIDIC_HASH']!,
+    });
+  } else {
+    print(
+      'WARNING: KANJIDIC version information not found in environment variables. '
+      'This may cause issues with future updates.',
+    );
+  }
+
  for (final c in characters) {
    // if (c.dictionaryReferences.any((e) =>
    //     c.dictionaryReferences
@@ -15,6 +33,24 @@ Future<void> seedKANJIDICData(List<Character> characters, Database db) async {
    //     }
    b.insert(KANJIDICTableNames.character, c.sqlValue);

+    if (c.grade != null) {
+      b.insert(KANJIDICTableNames.grade, {
+        'kanji': c.literal,
+        'grade': c.grade!,
+      });
+    }
+
+    if (c.frequency != null) {
+      b.insert(KANJIDICTableNames.frequency, {
+        'kanji': c.literal,
+        'frequency': c.frequency!,
+      });
+    }
+
+    if (c.jlpt != null) {
+      b.insert(KANJIDICTableNames.jlpt, {'kanji': c.literal, 'jlpt': c.jlpt!});
+    }
+
    for (final n in c.radicalName) {
      assert(c.radical != null, 'Radical name without radical');
      b.insert(
--- a/lib/_data_ingestion/radkfile/parser.dart
+++ b/lib/_data_ingestion/radkfile/parser.dart
@@ -1,7 +1,7 @@
 import 'dart:io';

 Iterable<String> parseRADKFILEBlocks(File radkfile) {
-  final String content = File('data/tmp/radkfile_utf8').readAsStringSync();
+  final String content = radkfile.readAsStringSync();

  final Iterable<String> blocks = content
      .replaceAll(RegExp(r'^#.*$'), '')
--- a/lib/_data_ingestion/radkfile/seed_data.dart
+++ b/lib/_data_ingestion/radkfile/seed_data.dart
@@ -1,9 +1,26 @@
+import 'dart:io';
+
 import 'package:jadb/table_names/radkfile.dart';
 import 'package:sqflite_common/sqlite_api.dart';

 Future<void> seedRADKFILEData(Iterable<String> blocks, Database db) async {
  final b = db.batch();

+  if (Platform.environment['RADKFILE_VERSION'] != null &&
+      Platform.environment['RADKFILE_DATE'] != null &&
+      Platform.environment['RADKFILE_HASH'] != null) {
+    b.insert(RADKFILETableNames.version, {
+      'version': Platform.environment['RADKFILE_VERSION']!,
+      'date': Platform.environment['RADKFILE_DATE']!,
+      'hash': Platform.environment['RADKFILE_HASH']!,
+    });
+  } else {
+    print(
+      'WARNING: RADKFILE version information not found in environment variables. '
+      'This may cause issues with future updates.',
+    );
+  }
+
  for (final block in blocks) {
    final String radical = block[1];
    final List<String> kanjiList =
--- a/lib/_data_ingestion/seed_database.dart
+++ b/lib/_data_ingestion/seed_database.dart
@@ -24,7 +24,7 @@ Future<void> seedData(Database db) async {

 Future<void> parseAndSeedDataFromJMdict(Database db) async {
  print('[JMdict] Reading file content...');
-  final String rawXML = File('data/tmp/JMdict.xml').readAsStringSync();
+  final String rawXML = File('data/JMdict.xml').readAsStringSync();

  print('[JMdict] Parsing XML tags...');
  final XmlElement root = XmlDocument.parse(rawXML).getElement('JMdict')!;
@@ -38,7 +38,7 @@ Future<void> parseAndSeedDataFromJMdict(Database db) async {

 Future<void> parseAndSeedDataFromKANJIDIC(Database db) async {
  print('[KANJIDIC2] Reading file...');
-  final String rawXML = File('data/tmp/kanjidic2.xml').readAsStringSync();
+  final String rawXML = File('data/kanjidic2.xml').readAsStringSync();

  print('[KANJIDIC2] Parsing XML...');
  final XmlElement root = XmlDocument.parse(rawXML).getElement('kanjidic2')!;
@@ -52,7 +52,7 @@ Future<void> parseAndSeedDataFromKANJIDIC(Database db) async {

 Future<void> parseAndSeedDataFromRADKFILE(Database db) async {
  print('[RADKFILE] Reading file...');
-  final File raw = File('data/tmp/RADKFILE');
+  final File raw = File('data/RADKFILE');

  print('[RADKFILE] Parsing content...');
  final blocks = parseRADKFILEBlocks(raw);
--- a/lib/_data_ingestion/tanos-jlpt/csv_parser.dart
+++ b/lib/_data_ingestion/tanos-jlpt/csv_parser.dart
@@ -3,18 +3,18 @@ import 'dart:io';

 import 'package:csv/csv.dart';
 import 'package:jadb/_data_ingestion/tanos-jlpt/objects.dart';
-import 'package:xml/xml_events.dart';

 Future<List<JLPTRankedWord>> parseJLPTRankedWords(
  Map<String, File> files,
 ) async {
  final List<JLPTRankedWord> result = [];

-  final codec = CsvCodec(
+  final codec = Csv(
    fieldDelimiter: ',',
    lineDelimiter: '\n',
    quoteMode: QuoteMode.strings,
    escapeCharacter: '\\',
+    parseHeaders: false,
  );

  for (final entry in files.entries) {
@@ -29,7 +29,6 @@ Future<List<JLPTRankedWord>> parseJLPTRankedWords(
        .openRead()
        .transform(utf8.decoder)
        .transform(codec.decoder)
-        .flatten()
        .map((row) {
          if (row.length != 3) {
            throw Exception('Invalid line in $jlptLevel: $row');
--- a/lib/_data_ingestion/tanos-jlpt/seed_data.dart
+++ b/lib/_data_ingestion/tanos-jlpt/seed_data.dart
@@ -1,3 +1,5 @@
+import 'dart:io';
+
 import 'package:jadb/table_names/tanos_jlpt.dart';
 import 'package:sqflite_common/sqlite_api.dart';

@@ -7,6 +9,21 @@ Future<void> seedTanosJLPTData(
 ) async {
  final Batch b = db.batch();

+  if (Platform.environment['TANOS_JLPT_VERSION'] != null &&
+      Platform.environment['TANOS_JLPT_DATE'] != null &&
+      Platform.environment['TANOS_JLPT_HASH'] != null) {
+    b.insert(TanosJLPTTableNames.version, {
+      'version': Platform.environment['TANOS_JLPT_VERSION']!,
+      'date': Platform.environment['TANOS_JLPT_DATE']!,
+      'hash': Platform.environment['TANOS_JLPT_HASH']!,
+    });
+  } else {
+    print(
+      'WARNING: Tanos JLPT version information not found in environment variables. '
+      'This may cause issues with future updates.',
+    );
+  }
+
  for (final jlptLevel in resolvedEntries.entries) {
    final level = jlptLevel.key;
    final entryIds = jlptLevel.value;
--- a/lib/models/word_search/word_search_result.dart
+++ b/lib/models/word_search/word_search_result.dart
@@ -1,5 +1,6 @@
 import 'package:jadb/models/common/jlpt_level.dart';
 import 'package:jadb/models/jmdict/jmdict_kanji_info.dart';
+import 'package:jadb/models/jmdict/jmdict_misc.dart';
 import 'package:jadb/models/jmdict/jmdict_reading_info.dart';
 import 'package:jadb/models/word_search/word_search_match_span.dart';
 import 'package:jadb/models/word_search/word_search_ruby.dart';
@@ -45,6 +46,13 @@ class WordSearchResult {
  /// the original searchword.
  List<WordSearchMatchSpan>? matchSpans;

+  /// Whether the first item in [japanese] contains kanji that likely is rare.
+  bool get hasUnusualKanji =>
+      (japanese.first.furigana != null &&
+          kanjiInfo[japanese.first.base] == JMdictKanjiInfo.rK) ||
+      senses.where((sense) => sense.misc.contains(JMdictMisc.onlyKana)).length >
+          (senses.length / 2);
+
  /// All contents of [japanese], transliterated to romaji
  List<String> get romaji => japanese
      .map((word) => transliterateKanaToLatin(word.furigana ?? word.base))
--- a/lib/search/kanji_search.dart
+++ b/lib/search/kanji_search.dart
@@ -8,10 +8,21 @@ import 'package:sqflite_common/sqflite.dart';
 Future<List<Map<String, Object?>>> _charactersQuery(
  DatabaseExecutor connection,
  String kanji,
-) => connection.query(
-  KANJIDICTableNames.character,
-  where: 'literal = ?',
-  whereArgs: [kanji],
+) => connection.rawQuery(
+  '''
+    SELECT
+      "${KANJIDICTableNames.character}"."literal",
+      "${KANJIDICTableNames.character}"."strokeCount",
+      "${KANJIDICTableNames.grade}"."grade",
+      "${KANJIDICTableNames.jlpt}"."jlpt",
+      "${KANJIDICTableNames.frequency}"."frequency"
+    FROM "${KANJIDICTableNames.character}"
+    LEFT JOIN "${KANJIDICTableNames.grade}" ON "${KANJIDICTableNames.character}"."literal" = "${KANJIDICTableNames.grade}"."kanji"
+    LEFT JOIN "${KANJIDICTableNames.jlpt}" ON "${KANJIDICTableNames.character}"."literal" = "${KANJIDICTableNames.jlpt}"."kanji"
+    LEFT JOIN "${KANJIDICTableNames.frequency}" ON "${KANJIDICTableNames.character}"."literal" = "${KANJIDICTableNames.frequency}"."kanji"
+    WHERE "literal" = ?
+  ''',
+  [kanji],
 );

 Future<List<Map<String, Object?>>> _codepointsQuery(
--- a/lib/search/word_search/data_query.dart
+++ b/lib/search/word_search/data_query.dart
@@ -69,7 +69,7 @@ Future<List<Map<String, Object?>>> _readingelementsQuery(
  JMdictTableNames.readingElement,
  where: 'entryId IN (${List.filled(entryIds.length, '?').join(',')})',
  whereArgs: entryIds,
-  orderBy: 'orderNum',
+  orderBy: 'elementId',
 );

 Future<List<Map<String, Object?>>> _kanjielementsQuery(
@@ -79,7 +79,7 @@ Future<List<Map<String, Object?>>> _kanjielementsQuery(
  JMdictTableNames.kanjiElement,
  where: 'entryId IN (${List.filled(entryIds.length, '?').join(',')})',
  whereArgs: entryIds,
-  orderBy: 'orderNum',
+  orderBy: 'elementId',
 );

 Future<List<Map<String, Object?>>> _jlpttagsQuery(
@@ -216,19 +216,43 @@ Future<List<Map<String, Object?>>> _sensepossQuery(
 Future<List<Map<String, Object?>>> _senserestrictedtokanjisQuery(
  DatabaseExecutor connection,
  List<int> senseIds,
-) => connection.query(
-  JMdictTableNames.senseRestrictedToKanji,
-  where: 'senseId IN (${List.filled(senseIds.length, '?').join(',')})',
-  whereArgs: senseIds,
+) => connection.rawQuery(
+  """
+      SELECT
+        "${JMdictTableNames.senseRestrictedToKanji}".senseId,
+        "${JMdictTableNames.senseRestrictedToKanji}".kanjiElementId,
+        "${JMdictTableNames.kanjiElement}".reading
+      FROM "${JMdictTableNames.senseRestrictedToKanji}"
+      JOIN "${JMdictTableNames.kanjiElement}"
+        ON "${JMdictTableNames.senseRestrictedToKanji}"."kanjiElementId" = "${JMdictTableNames.kanjiElement}"."elementId"
+      WHERE
+        "senseId" IN (${List.filled(senseIds.length, '?').join(',')})
+      ORDER BY
+        "${JMdictTableNames.senseRestrictedToKanji}"."senseId",
+        "${JMdictTableNames.senseRestrictedToKanji}"."kanjiElementId"
+  """,
+  [...senseIds],
 );

 Future<List<Map<String, Object?>>> _senserestrictedtoreadingsQuery(
  DatabaseExecutor connection,
  List<int> senseIds,
-) => connection.query(
-  JMdictTableNames.senseRestrictedToReading,
-  where: 'senseId IN (${List.filled(senseIds.length, '?').join(',')})',
-  whereArgs: senseIds,
+) => connection.rawQuery(
+  """
+      SELECT
+        "${JMdictTableNames.senseRestrictedToReading}".senseId,
+        "${JMdictTableNames.senseRestrictedToReading}".readingElementId,
+        "${JMdictTableNames.readingElement}".reading
+      FROM "${JMdictTableNames.senseRestrictedToReading}"
+      JOIN "${JMdictTableNames.readingElement}"
+        ON "${JMdictTableNames.senseRestrictedToReading}"."readingElementId" = "${JMdictTableNames.readingElement}"."elementId"
+      WHERE
+        "senseId" IN (${List.filled(senseIds.length, '?').join(',')})
+      ORDER BY
+        "${JMdictTableNames.senseRestrictedToReading}"."senseId",
+        "${JMdictTableNames.senseRestrictedToReading}"."readingElementId"
+  """,
+  [...senseIds],
 );

 Future<List<Map<String, Object?>>> _examplesentencesQuery(
--- a/lib/search/word_search/entry_id_query.dart
+++ b/lib/search/word_search/entry_id_query.dart
@@ -68,25 +68,25 @@ String _filterFTSSensitiveCharacters(String word) {
              "$tableName"."entryId",
              100
              + (("${tableName}FTS"."reading" = ?) * 10000)
-              + "JMdict_EntryScore"."score"
+              + (("$tableName"."orderNum" = 0) * 20)
+              + COALESCE("JMdict_EntryScore"."score", 0)
              AS "score"
            FROM "${tableName}FTS"
            JOIN "$tableName" USING ("elementId")
-            JOIN "JMdict_EntryScore" USING ("elementId")
+            LEFT JOIN "JMdict_EntryScore" USING ("elementId")
            WHERE "${tableName}FTS"."reading" MATCH ? || '*'
-              AND "JMdict_EntryScore"."type" = '${tableName == JMdictTableNames.kanjiElement ? 'k' : 'r'}'
          ),
          non_fts_results AS (
            SELECT DISTINCT
              "$tableName"."entryId",
              50
-              + "JMdict_EntryScore"."score"
+              + (("$tableName"."orderNum" = 0) * 20)
+              + COALESCE("JMdict_EntryScore"."score", 0)
              AS "score"
            FROM "$tableName"
-            JOIN "JMdict_EntryScore" USING ("elementId")
+            LEFT JOIN "JMdict_EntryScore" USING ("elementId")
            WHERE "reading" LIKE '%' || ? || '%'
              AND "$tableName"."entryId" NOT IN (SELECT "entryId" FROM "fts_results")
-              AND "JMdict_EntryScore"."type" = '${tableName == JMdictTableNames.kanjiElement ? 'k' : 'r'}'
          )

        SELECT ${countOnly ? 'COUNT(DISTINCT "entryId") AS count' : '"entryId", MAX("score") AS "score"'}
@@ -198,16 +198,16 @@ Future<List<ScoredEntryId>> _queryEnglish(
    '''
    SELECT
      "${JMdictTableNames.sense}"."entryId",
-      MAX("JMdict_EntryScore"."score")
-      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1 AND "${JMdictTableNames.sense}"."orderNum" = 1) * 50)
-      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1 AND "${JMdictTableNames.sense}"."orderNum" = 2) * 30)
-      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1) * 20)
+      COALESCE(MAX("JMdict_EntryScore"."score"), 0)
+      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1 AND "${JMdictTableNames.sense}"."orderNum" = 0) * 50)
+      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1 AND "${JMdictTableNames.sense}"."orderNum" = 1) * 30)
+      + (("${JMdictTableNames.senseGlossary}"."phrase" = ?1 AND "${JMdictTableNames.sense}"."orderNum" > 1) * 20)
      as "score"
    FROM "${JMdictTableNames.senseGlossary}"
    JOIN "${JMdictTableNames.sense}" USING ("senseId")
-    JOIN "JMdict_EntryScore" USING ("entryId")
+    LEFT JOIN "JMdict_EntryScore" USING ("entryId")
    WHERE "${JMdictTableNames.senseGlossary}"."phrase" LIKE ?2
-    GROUP BY "JMdict_EntryScore"."entryId"
+    GROUP BY "${JMdictTableNames.sense}"."entryId"
    ORDER BY
      "score" DESC,
      "${JMdictTableNames.sense}"."entryId" ASC
@@ -215,7 +215,7 @@ Future<List<ScoredEntryId>> _queryEnglish(
    ${offset != null ? 'OFFSET ?4' : ''}
  '''
        .trim(),
-    [word, '%${word.replaceAll('%', '')}%', if (pageSize != null) pageSize, if (offset != null) offset],
+    [word, '%${word.replaceAll('%', '')}%', ?pageSize, ?offset],
  );

  return result
--- a/lib/search/word_search/regrouping.dart
+++ b/lib/search/word_search/regrouping.dart
@@ -318,7 +318,7 @@ List<WordSearchSense> _regroupSenses({
          .map((e) => e['reading'] as String)
          .toList(),
      restrictedToKanji: restrictedToKanjis
-          .map((e) => e['kanji'] as String)
+          .map((e) => e['reading'] as String)
          .toList(),
      fields: fields
          .map((e) => JMdictField.fromId(e['field'] as String))
--- a/lib/table_names/jmdict.dart
+++ b/lib/table_names/jmdict.dart
@@ -11,6 +11,7 @@ abstract class JMdictTableNames {
  static const String senseDialect = 'JMdict_SenseDialect';
  static const String senseField = 'JMdict_SenseField';
  static const String senseGlossary = 'JMdict_SenseGlossary';
+  static const String senseGlossaryType = 'JMdict_SenseGlossaryType';
  static const String senseInfo = 'JMdict_SenseInfo';
  static const String senseMisc = 'JMdict_SenseMisc';
  static const String sensePOS = 'JMdict_SensePOS';
@@ -33,6 +34,7 @@ abstract class JMdictTableNames {
    senseDialect,
    senseField,
    senseGlossary,
+    senseGlossaryType,
    senseInfo,
    senseMisc,
    sensePOS,
--- a/lib/table_names/kanjidic.dart
+++ b/lib/table_names/kanjidic.dart
@@ -1,6 +1,9 @@
 abstract class KANJIDICTableNames {
  static const String version = 'KANJIDIC_Version';
  static const String character = 'KANJIDIC_Character';
+  static const String grade = 'KANJIDIC_Grade';
+  static const String frequency = 'KANJIDIC_Frequency';
+  static const String jlpt = 'KANJIDIC_JLPT';
  static const String radicalName = 'KANJIDIC_RadicalName';
  static const String codepoint = 'KANJIDIC_Codepoint';
  static const String radical = 'KANJIDIC_Radical';
@@ -20,6 +23,9 @@ abstract class KANJIDICTableNames {
  static Set<String> get allTables => {
    version,
    character,
+    grade,
+    frequency,
+    jlpt,
    radicalName,
    codepoint,
    radical,
--- a/migrations/0001_JMDict.sql
+++ b/migrations/0001_JMDict.sql
@@ -53,19 +53,18 @@ CREATE TABLE "JMdict_Entry" (

 CREATE TABLE "JMdict_KanjiElement" (
  "elementId" INTEGER PRIMARY KEY,
-  "entryId" INTEGER NOT NULL REFERENCES "JMdict_Entry"("entryId"),
-  "orderNum" INTEGER NOT NULL,
+  "entryId" INTEGER NOT NULL GENERATED ALWAYS AS ("elementId" / 100) STORED,
+  "orderNum" INTEGER NOT NULL GENERATED ALWAYS AS ("elementId" % 100) VIRTUAL,
  "reading" TEXT NOT NULL,
  "news" INTEGER CHECK ("news" BETWEEN 1 AND 2),
  "ichi" INTEGER CHECK ("ichi" BETWEEN 1 AND 2),
  "spec" INTEGER CHECK ("spec" BETWEEN 1 AND 2),
  "gai" INTEGER CHECK ("gai" BETWEEN 1 AND 2),
  "nf" INTEGER CHECK ("nf" BETWEEN 1 AND 48),
-  UNIQUE("entryId", "reading"),
-  UNIQUE("entryId", "orderNum")
+  FOREIGN KEY ("entryId") REFERENCES "JMdict_Entry"("entryId"),
+  UNIQUE("entryId", "reading")
 ) WITHOUT ROWID;

-CREATE INDEX "JMdict_KanjiElement_byEntryId_byOrderNum" ON "JMdict_KanjiElement"("entryId", "orderNum");
 CREATE INDEX "JMdict_KanjiElement_byReading" ON "JMdict_KanjiElement"("reading");

 CREATE TABLE "JMdict_KanjiElementInfo" (
@@ -78,8 +77,8 @@ CREATE TABLE "JMdict_KanjiElementInfo" (

 CREATE TABLE "JMdict_ReadingElement" (
  "elementId" INTEGER PRIMARY KEY,
-  "entryId" INTEGER NOT NULL REFERENCES "JMdict_Entry"("entryId"),
-  "orderNum" INTEGER NOT NULL,
+  "entryId" INTEGER NOT NULL GENERATED ALWAYS AS (("elementId" / 100) % 10000000) STORED,
+  "orderNum" INTEGER NOT NULL GENERATED ALWAYS AS ("elementId" % 100) VIRTUAL,
  "reading" TEXT NOT NULL,
  "readingDoesNotMatchKanji" BOOLEAN NOT NULL DEFAULT FALSE,
  "news" INTEGER CHECK ("news" BETWEEN 1 AND 2),
@@ -87,11 +86,10 @@ CREATE TABLE "JMdict_ReadingElement" (
  "spec" INTEGER CHECK ("spec" BETWEEN 1 AND 2),
  "gai" INTEGER CHECK ("gai" BETWEEN 1 AND 2),
  "nf" INTEGER CHECK ("nf" BETWEEN 1 AND 48),
-  UNIQUE("entryId", "reading"),
-  UNIQUE("entryId", "orderNum")
+  FOREIGN KEY ("entryId") REFERENCES "JMdict_Entry"("entryId"),
+  UNIQUE("entryId", "reading")
 ) WITHOUT ROWID;

-CREATE INDEX "JMdict_ReadingElement_byEntryId_byOrderNum" ON "JMdict_ReadingElement"("entryId", "orderNum");
 CREATE INDEX "JMdict_ReadingElement_byReading" ON "JMdict_ReadingElement"("reading");

 CREATE TABLE "JMdict_ReadingElementRestriction" (
@@ -110,27 +108,22 @@ CREATE TABLE "JMdict_ReadingElementInfo" (

 CREATE TABLE "JMdict_Sense" (
  "senseId" INTEGER PRIMARY KEY,
-  "entryId" INTEGER NOT NULL REFERENCES "JMdict_Entry"("entryId"),
-  "orderNum" INTEGER NOT NULL,
+  "entryId" INTEGER NOT NULL GENERATED ALWAYS AS ("senseId" / 100) STORED,
+  "orderNum" INTEGER NOT NULL GENERATED ALWAYS AS ("senseId" % 100) VIRTUAL,
+  FOREIGN KEY ("entryId") REFERENCES "JMdict_Entry"("entryId"),
  UNIQUE("entryId", "orderNum")
 );

-CREATE INDEX "JMdict_Sense_byEntryId_byOrderNum" ON "JMdict_Sense"("entryId", "orderNum");
-
 CREATE TABLE "JMdict_SenseRestrictedToKanji" (
-  "entryId" INTEGER NOT NULL,
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
-  "kanji" TEXT NOT NULL,
-  FOREIGN KEY ("entryId", "kanji") REFERENCES "JMdict_KanjiElement"("entryId", "reading"),
-  PRIMARY KEY ("entryId", "senseId", "kanji")
+  "kanjiElementId" INTEGER NOT NULL REFERENCES "JMdict_KanjiElement"("elementId"),
+  PRIMARY KEY ("senseId", "kanjiElementId")
 ) WITHOUT ROWID;

 CREATE TABLE "JMdict_SenseRestrictedToReading" (
-  "entryId" INTEGER NOT NULL,
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
-  "reading" TEXT NOT NULL,
-  FOREIGN KEY ("entryId", "reading") REFERENCES "JMdict_ReadingElement"("entryId", "reading"),
-  PRIMARY KEY ("entryId", "senseId", "reading")
+  "readingElementId" INTEGER NOT NULL REFERENCES "JMdict_ReadingElement"("elementId"),
+  PRIMARY KEY ("senseId", "readingElementId")
 ) WITHOUT ROWID;

 -- In order to add xrefs, you will need to have added the entry to xref to.
@@ -146,31 +139,41 @@ CREATE TABLE "JMdict_SenseRestrictedToReading" (
 CREATE TABLE "JMdict_SenseSeeAlso" (
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
  "xrefEntryId" INTEGER NOT NULL,
-  "seeAlsoReading" TEXT,
-  "seeAlsoKanji" TEXT,
  "seeAlsoSense" INTEGER,
  -- For some entries, the cross reference is ambiguous. This means that while the ingestion
  -- has determined some xrefEntryId, it is not guaranteed to be the correct one.
  "ambiguous" BOOLEAN NOT NULL DEFAULT FALSE,
-  FOREIGN KEY ("xrefEntryId", "seeAlsoKanji") REFERENCES "JMdict_KanjiElement"("entryId", "reading"),
-  FOREIGN KEY ("xrefEntryId", "seeAlsoReading") REFERENCES "JMdict_ReadingElement"("entryId", "reading"),
-  FOREIGN KEY ("xrefEntryId", "seeAlsoSense") REFERENCES "JMdict_Sense"("entryId", "orderNum"),
-  UNIQUE("senseId", "xrefEntryId", "seeAlsoReading", "seeAlsoKanji", "seeAlsoSense")
+
+  "seeAlsoSenseKey" INTEGER GENERATED ALWAYS AS (
+    CASE
+      WHEN "seeAlsoSense" IS NOT NULL THEN ("xrefEntryId" * 100) + "seeAlsoSense"
+      ELSE NULL
+    END
+  ) VIRTUAL,
+
+  FOREIGN KEY ("seeAlsoSenseKey") REFERENCES "JMdict_Sense"("senseId"),
+
+  PRIMARY KEY ("senseId", "xrefEntryId", "seeAlsoSense")
 );

 CREATE TABLE "JMdict_SenseAntonym" (
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
  "xrefEntryId" INTEGER NOT NULL,
-  "antonymReading" TEXT,
-  "antonymKanji" TEXT,
  "antonymSense" INTEGER,
  -- For some entries, the cross reference is ambiguous. This means that while the ingestion
  -- has determined some xrefEntryId, it is not guaranteed to be the correct one.
  "ambiguous" BOOLEAN NOT NULL DEFAULT FALSE,
-  FOREIGN KEY ("xrefEntryId", "antonymKanji") REFERENCES "JMdict_KanjiElement"("entryId", "reading"),
-  FOREIGN KEY ("xrefEntryId", "antonymReading") REFERENCES "JMdict_ReadingElement"("entryId", "reading"),
-  FOREIGN KEY ("xrefEntryId", "antonymSense") REFERENCES "JMdict_Sense"("entryId", "orderNum"),
-  UNIQUE("senseId", "xrefEntryId", "antonymReading", "antonymKanji", "antonymSense")
+
+  "antonymSenseKey" INTEGER GENERATED ALWAYS AS (
+    CASE
+      WHEN "antonymSense" IS NOT NULL THEN ("xrefEntryId" * 100) + "antonymSense"
+      ELSE NULL
+    END
+  ) VIRTUAL,
+
+  FOREIGN KEY ("antonymSenseKey") REFERENCES "JMdict_Sense"("senseId"),
+
+  PRIMARY KEY ("senseId", "xrefEntryId", "antonymSense")
 );

 -- These cross references are going to be mostly accessed from a sense
@@ -221,12 +224,20 @@ CREATE TABLE "JMdict_SenseDialect" (
 CREATE TABLE "JMdict_SenseGlossary" (
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
  "phrase" TEXT NOT NULL,
-  "language" CHAR(3) NOT NULL DEFAULT "eng",
-  "type" TEXT,
-  PRIMARY KEY ("senseId", "language", "phrase")
+  -- "language" CHAR(3) NOT NULL DEFAULT "eng",
+  -- PRIMARY KEY ("senseId", "language", "phrase")
+  PRIMARY KEY ("senseId", "phrase")
 ) WITHOUT ROWID;

-CREATE INDEX "JMdict_SenseGlossary_byPhrase" ON JMdict_SenseGlossary("phrase");
+-- CREATE INDEX "JMdict_SenseGlossary_byPhrase" ON JMdict_SenseGlossary("phrase");
+
+CREATE TABLE "JMdict_SenseGlossaryType" (
+  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
+  "phrase" TEXT NOT NULL,
+  "type" TEXT NOT NULL,
+  PRIMARY KEY ("senseId", "phrase", "type"),
+  FOREIGN KEY ("senseId", "phrase") REFERENCES "JMdict_SenseGlossary"("senseId", "phrase")
+) WITHOUT ROWID;

 CREATE TABLE "JMdict_SenseInfo" (
  "senseId" INTEGER NOT NULL REFERENCES "JMdict_Sense"("senseId"),
--- a/migrations/0005_JMDict_search_index_tables.sql
+++ b/migrations/0005_JMDict_search_index_tables.sql
@@ -1,26 +1,28 @@
 CREATE TABLE "JMdict_EntryScore" (
-  "type" CHAR(1) NOT NULL CHECK ("type" IN ('r', 'k')),
-  "entryId" INTEGER NOT NULL REFERENCES "JMdict_Entry"("entryId"),
-  "elementId" INTEGER NOT NULL,
+  "elementId" INTEGER PRIMARY KEY,
  "score" INTEGER NOT NULL DEFAULT 0,
  "common" BOOLEAN NOT NULL DEFAULT FALSE,
-  PRIMARY KEY ("type", "elementId")
+
+  "entryId" INTEGER NOT NULL GENERATED ALWAYS AS (("elementId" / 100) % 10000000) STORED,
+  "type" CHAR(1) NOT NULL GENERATED ALWAYS AS (CASE
+    WHEN "elementId" / 1000000000 = 0 THEN 'k'
+    ELSE 'r'
+  END) VIRTUAL,
+
+  FOREIGN KEY ("entryId") REFERENCES "JMdict_Entry"("entryId")
 ) WITHOUT ROWID;

 CREATE INDEX "JMdict_EntryScore_byElementId_byScore" ON "JMdict_EntryScore"("elementId", "score");
 CREATE INDEX "JMdict_EntryScore_byScore" ON "JMdict_EntryScore"("score");
-CREATE INDEX "JMdict_EntryScore_byCommon" ON "JMdict_EntryScore"("common");

-CREATE INDEX "JMdict_EntryScore_byType_byElementId_byScore" ON "JMdict_EntryScore"("type", "elementId", "score");
-CREATE INDEX "JMdict_EntryScore_byType_byScore" ON "JMdict_EntryScore"("type", "score");
-CREATE INDEX "JMdict_EntryScore_byType_byCommon" ON "JMdict_EntryScore"("type", "common");
+CREATE INDEX "JMdict_EntryScore_byElementId_byCommon" ON "JMdict_EntryScore"("elementId", "common");
+CREATE INDEX "JMdict_EntryScore_byCommon" ON "JMdict_EntryScore"("common");

 -- NOTE: these views are deduplicated in order not to perform an unnecessary
 --       UNION on every trigger

 CREATE VIEW "JMdict_EntryScoreView_Reading" AS
 SELECT
-  'r' AS "type",
  "JMdict_ReadingElement"."entryId",
  "JMdict_ReadingElement"."elementId",
  (
@@ -44,7 +46,7 @@ SELECT
  + (("spec" IS 2) * 5)
  + (("gai" IS 1) * 10)
  + (("gai" IS 2) * 5)
-  + (("orderNum" IS 1) * 20)
+  -- + (("orderNum" IS 0) * 20)
  - (substr(COALESCE("JMdict_JLPTTag"."jlptLevel", 'N0'), 2) * -5)
  AS "score"
 FROM "JMdict_ReadingElement"
@@ -52,7 +54,6 @@ LEFT JOIN "JMdict_JLPTTag" USING ("entryId");

 CREATE VIEW "JMdict_EntryScoreView_Kanji" AS
 SELECT
-  'k' AS "type",
  "JMdict_KanjiElement"."entryId",
  "JMdict_KanjiElement"."elementId",
  (
@@ -76,7 +77,7 @@ SELECT
  + (("spec" IS 2) * 5)
  + (("gai" IS 1) * 10)
  + (("gai" IS 2) * 5)
-  + (("orderNum" IS 1) * 20)
+  -- + (("orderNum" IS 0) * 20)
  - (substr(COALESCE("JMdict_JLPTTag"."jlptLevel", 'N0'), 2) * -5)
  AS "score"
 FROM "JMdict_KanjiElement"
@@ -96,19 +97,18 @@ CREATE TRIGGER "JMdict_EntryScore_Insert_JMdict_ReadingElement"
 AFTER INSERT ON "JMdict_ReadingElement"
 BEGIN
  INSERT INTO "JMdict_EntryScore" (
-    "type",
-    "entryId",
    "elementId",
    "score",
    "common"
  )
-  SELECT "type", "entryId", "elementId", "score", "common"
+  SELECT "elementId", "score", "common"
  FROM "JMdict_EntryScoreView_Reading"
-  WHERE "elementId" = NEW."elementId";
+  WHERE "elementId" = NEW."elementId"
+    AND "score" > 0;
 END;

 CREATE TRIGGER "JMdict_EntryScore_Update_JMdict_ReadingElement"
-AFTER UPDATE OF "news", "ichi", "spec", "gai", "nf", "orderNum"
+AFTER UPDATE OF "news", "ichi", "spec", "gai", "nf", "elementId"
 ON "JMdict_ReadingElement"
 BEGIN
  UPDATE "JMdict_EntryScore"
@@ -117,14 +117,17 @@ BEGIN
    "common" = "JMdict_EntryScoreView_Reading"."common"
  FROM "JMdict_EntryScoreView_Reading"
  WHERE "elementId" = NEW."elementId";
+
+  DELETE FROM "JMdict_EntryScore"
+  WHERE "elementId" = NEW."elementId"
+    AND "score" <= 0;
 END;

 CREATE TRIGGER "JMdict_EntryScore_Delete_JMdict_ReadingElement"
 AFTER DELETE ON "JMdict_ReadingElement"
 BEGIN
  DELETE FROM "JMdict_EntryScore"
-  WHERE "type" = 'r'
-    AND "elementId" = OLD."elementId";
+  WHERE "elementId" = OLD."elementId";
 END;

 --- JMdict_KanjiElement triggers
@@ -133,19 +136,18 @@ CREATE TRIGGER "JMdict_EntryScore_Insert_JMdict_KanjiElement"
 AFTER INSERT ON "JMdict_KanjiElement"
 BEGIN
  INSERT INTO "JMdict_EntryScore" (
-    "type",
-    "entryId",
    "elementId",
    "score",
    "common"
  )
-  SELECT "type", "entryId", "elementId", "score", "common"
+  SELECT "elementId", "score", "common"
  FROM "JMdict_EntryScoreView_Kanji"
-  WHERE "elementId" = NEW."elementId";
+  WHERE "elementId" = NEW."elementId"
+    AND "score" > 0;
 END;

 CREATE TRIGGER "JMdict_EntryScore_Update_JMdict_KanjiElement"
-AFTER UPDATE OF "news", "ichi", "spec", "gai", "nf", "orderNum"
+AFTER UPDATE OF "news", "ichi", "spec", "gai", "nf", "elementId"
 ON "JMdict_KanjiElement"
 BEGIN
  UPDATE "JMdict_EntryScore"
@@ -154,14 +156,17 @@ BEGIN
    "common" = "JMdict_EntryScoreView_Kanji"."common"
  FROM "JMdict_EntryScoreView_Kanji"
  WHERE "elementId" = NEW."elementId";
+
+  DELETE FROM "JMdict_EntryScore"
+  WHERE "elementId" = NEW."elementId"
+    AND "score" <= 0;
 END;

 CREATE TRIGGER "JMdict_EntryScore_Delete_JMdict_KanjiElement"
 AFTER DELETE ON "JMdict_KanjiElement"
 BEGIN
  DELETE FROM "JMdict_EntryScore"
-  WHERE "type" = 'k'
-    AND "elementId" = OLD."elementId";
+  WHERE "elementId" = OLD."elementId";
 END;

 --- JMdict_JLPTTag triggers
@@ -204,4 +209,8 @@ BEGIN
  WHERE "JMdict_EntryScoreView"."entryId" = OLD."entryId"
    AND "JMdict_EntryScore"."entryId" = OLD."entryId"
    AND "JMdict_EntryScoreView"."elementId" = "JMdict_EntryScore"."elementId";
+
+  DELETE FROM "JMdict_EntryScore"
+  WHERE "elementId" = OLD."elementId"
+    AND "score" <= 0;
 END;
--- a/migrations/0006_RADKFILE.sql
+++ b/migrations/0006_RADKFILE.sql
@@ -18,4 +18,3 @@ CREATE TABLE "RADKFILE" (
 ) WITHOUT ROWID;

 CREATE INDEX "RADK" ON "RADKFILE"("radical");
-CREATE INDEX "KRAD" ON "RADKFILE"("kanji");
--- a/migrations/0007_KANJIDIC2.sql
+++ b/migrations/0007_KANJIDIC2.sql
@@ -19,6 +19,21 @@ CREATE TABLE "KANJIDIC_Character" (
  "jlpt" INTEGER
 ) WITHOUT ROWID;

+CREATE TABLE "KANJIDIC_Grade" (
+  "kanji" CHAR(1) NOT NULL PRIMARY KEY REFERENCES "KANJIDIC_Character"("literal"),
+  "grade" INTEGER NOT NULL CHECK ("grade" BETWEEN 1 AND 10)
+) WITHOUT ROWID;
+
+CREATE TABLE "KANJIDIC_Frequency" (
+  "kanji" CHAR(1) NOT NULL PRIMARY KEY REFERENCES "KANJIDIC_Character"("literal"),
+  "frequency" INTEGER NOT NULL
+) WITHOUT ROWID;
+
+CREATE TABLE "KANJIDIC_JLPT" (
+  "kanji" CHAR(1) NOT NULL PRIMARY KEY REFERENCES "KANJIDIC_Character"("literal"),
+  "jlpt" INTEGER NOT NULL CHECK ("jlpt" BETWEEN 1 AND 5)
+) WITHOUT ROWID;
+
 CREATE TABLE "KANJIDIC_Codepoint" (
  "kanji" CHAR(1) NOT NULL REFERENCES "KANJIDIC_Character"("literal"),
  "type" VARCHAR(6) NOT NULL CHECK ("type" IN ('jis208', 'jis212', 'jis213', 'ucs')),
--- a/migrations/0008_XREF__KANJIDIC_Radical___RADKFILE.sql
+++ b/migrations/0008_XREF__KANJIDIC_Radical___RADKFILE.sql
@@ -4,7 +4,6 @@ CREATE TABLE "XREF__KANJIDIC_Radical__RADKFILE"(
  PRIMARY KEY ("radicalId", "radicalSymbol")
 ) WITHOUT ROWID;

-CREATE INDEX "XREF__KANJIDIC_Radical__RADKFILE__byRadicalId" ON "XREF__KANJIDIC_Radical__RADKFILE"("radicalId");
 CREATE INDEX "XREF__KANJIDIC_Radical__RADKFILE__byRadicalSymbol" ON "XREF__KANJIDIC_Radical__RADKFILE"("radicalSymbol");

 /* Source: https://ctext.org/kangxi-zidian */
--- a/migrations/0009_XREF__JMdict_KanjiElement__KANJIDIC_Character.sql
+++ b/migrations/0009_XREF__JMdict_KanjiElement__KANJIDIC_Character.sql
@@ -1,10 +1,10 @@
-CREATE TABLE "XREF__JMdict_KanjiElement__KANJIDIC_Character"(
-  "entryId" INTEGER NOT NULL,
-  "reading" TEXT NOT NULL,
-  "kanji" CHAR(1) NOT NULL REFERENCES "KANJIDIC_Character"("literal"),
-  PRIMARY KEY ("entryId", "reading", "kanji"),
-  FOREIGN KEY ("entryId", "reading") REFERENCES "JMdict_KanjiElement"("entryId", "reading")
-) WITHOUT ROWID;
+-- CREATE TABLE "XREF__JMdict_KanjiElement__KANJIDIC_Character"(
+--   "entryId" INTEGER NOT NULL,
+--   "reading" TEXT NOT NULL,
+--   "kanji" CHAR(1) NOT NULL REFERENCES "KANJIDIC_Character"("literal"),
+--   PRIMARY KEY ("entryId", "reading", "kanji"),
+--   FOREIGN KEY ("entryId", "reading") REFERENCES "JMdict_KanjiElement"("entryId", "reading")
+-- ) WITHOUT ROWID;

-CREATE INDEX "XREF__JMdict_KanjiElement__KANJIDIC_Character__byEntryId_byReading" ON "XREF__JMdict_KanjiElement__KANJIDIC_Character"("entryId", "reading");
-CREATE INDEX "XREF__JMdict_KanjiElement__KANJIDIC_Character__byKanji" ON "XREF__JMdict_KanjiElement__KANJIDIC_Character"("kanji");
+-- CREATE INDEX "XREF__JMdict_KanjiElement__KANJIDIC_Character__byEntryId_byReading" ON "XREF__JMdict_KanjiElement__KANJIDIC_Character"("entryId", "reading");
+-- CREATE INDEX "XREF__JMdict_KanjiElement__KANJIDIC_Character__byKanji" ON "XREF__JMdict_KanjiElement__KANJIDIC_Character"("kanji");
--- a/migrations/0010_Views.sql
+++ b/migrations/0010_Views.sql
@@ -32,9 +32,9 @@ SELECT
    THEN "JMdict_ReadingElement"."reading"
    ELSE NULL
  END AS "furigana",
-  COALESCE("JMdict_KanjiElement"."orderNum", 1)
+  COALESCE("JMdict_KanjiElement"."orderNum", 0)
      + "JMdict_ReadingElement"."orderNum"
-      = 2
+      = 0
  AS "isFirst",
  "JMdict_KanjiElement"."orderNum" AS "kanjiOrderNum",
  "JMdict_ReadingElement"."orderNum" AS "readingOrderNum"
@@ -65,9 +65,7 @@ JOIN "JMdict_KanjiElement"
  ON "JMdict_KanjiElementFTS"."entryId" = "JMdict_KanjiElement"."entryId"
    AND "JMdict_KanjiElementFTS"."reading" LIKE '%' || "JMdict_KanjiElement"."reading"
 JOIN "JMdict_EntryScore"
-  ON "JMdict_EntryScore"."type" = 'k'
-    AND "JMdict_KanjiElement"."entryId" = "JMdict_EntryScore"."entryId"
-    AND "JMdict_KanjiElement"."reading" = "JMdict_EntryScore"."reading"
+  ON "JMdict_EntryScore"."elementId" = "JMdict_KanjiElement"."elementId"
 WHERE "JMdict_EntryScore"."common" = 1;


@@ -77,13 +75,7 @@ SELECT DISTINCT "radical" FROM "RADKFILE";
 CREATE VIEW "JMdict_CombinedEntryScore"
 AS
 SELECT
-  CASE
-    WHEN "JMdict_EntryScore"."type" = 'k'
-      THEN (SELECT entryId FROM "JMdict_KanjiElement" WHERE "elementId" = "JMdict_EntryScore"."elementId")
-    WHEN "JMdict_EntryScore"."type" = 'r'
-      THEN (SELECT entryId FROM "JMdict_ReadingElement" WHERE "elementId" = "JMdict_EntryScore"."elementId")
-  END AS "entryId",
  MAX("JMdict_EntryScore"."score") AS "score",
  MAX("JMdict_EntryScore"."common") AS "common"
 FROM "JMdict_EntryScore"
-GROUP BY "entryId";
+GROUP BY "JMdict_EntryScore"."entryId";
--- a/nix/database.nix
+++ b/nix/database.nix
@@ -6,6 +6,7 @@
  jmdict,
  radkfile,
  kanjidic2,
+  tanos-jlpt,
  sqlite,
  wal ? false,
 }:
@@ -18,13 +19,32 @@ stdenvNoCC.mkDerivation {
    sqlite
  ];

+  env = {
+    JMDICT_VERSION = jmdict.version;
+    JMDICT_DATE = jmdict.date;
+    JMDICT_HASH = jmdict.hash;
+
+    KANJIDIC_VERSION = kanjidic2.version;
+    KANJIDIC_DATE = kanjidic2.date;
+    KANJIDIC_HASH = kanjidic2.hash;
+
+    RADKFILE_VERSION = radkfile.version;
+    RADKFILE_DATE = radkfile.date;
+    RADKFILE_HASH = radkfile.hash;
+
+    TANOS_JLPT_VERSION = tanos-jlpt.version;
+    TANOS_JLPT_DATE = tanos-jlpt.date;
+    TANOS_JLPT_HASH = tanos-jlpt.hash;
+  };
+
  buildPhase = ''
    runHook preBuild

-    mkdir -p data/tmp
-    ln -s "${jmdict}"/* data/tmp
-    ln -s "${radkfile}"/* data/tmp
-    ln -s "${kanjidic2}"/* data/tmp
+    mkdir -p data
+    ln -s '${jmdict}'/* data/
+    ln -s '${kanjidic2}'/* data/
+    ln -s '${radkfile}'/* data/
+    ln -s '${tanos-jlpt}' data/tanos-jlpt

    for migration in migrations/*.sql; do
      sqlite3 jadb.sqlite < "$migration"
--- a/nix/jmdict.nix
+++ b/nix/jmdict.nix
@@ -1,46 +0,0 @@
-{
-  stdenvNoCC,
-  jmdict-src,
-  jmdict-with-examples-src,
-  xmlformat,
-  gzip,
-  edrdgMetadata,
-}:
-stdenvNoCC.mkDerivation {
-  name = "jmdict";
-
-  dontUnpack = true;
-  srcs = [
-    jmdict-src
-    jmdict-with-examples-src
-  ];
-
-  nativeBuildInputs = [
-    gzip
-    xmlformat
-  ];
-
-  buildPhase = ''
-    runHook preBuild
-
-    gzip -dkc "${jmdict-src}" > JMdict.xml
-    gzip -dkc "${jmdict-with-examples-src}" > JMdict_with_examples.xml
-    xmlformat -i JMdict.xml
-    xmlformat -i JMdict_with_examples.xml
-
-    runHook postBuild
-  '';
-
-  installPhase = ''
-    runHook preInstall
-
-    install -Dt "$out" JMdict.xml JMdict_with_examples.xml
-
-    runHook postInstall
-  '';
-
-  meta = edrdgMetadata // {
-    description = "A Japanese-Multilingual Dictionary providing lexical data for japanese words";
-    homepage = "https://www.edrdg.org/jmdict/j_jmdict.html";
-  };
-}
--- a/nix/kanjidic2.nix
+++ b/nix/kanjidic2.nix
@@ -1,40 +0,0 @@
-{
-  stdenvNoCC,
-  kanjidic2-src,
-  xmlformat,
-  gzip,
-  edrdgMetadata,
-}:
-stdenvNoCC.mkDerivation {
-  name = "kanjidic2";
-
-  src = kanjidic2-src;
-  dontUnpack = true;
-
-  nativeBuildInputs = [
-    gzip
-    xmlformat
-  ];
-
-  buildPhase = ''
-    runHook preBuild
-
-    gzip -dkc "${kanjidic2-src}" > kanjidic2.xml
-    xmlformat -i kanjidic2.xml
-
-    runHook postBuild
-  '';
-
-  installPhase = ''
-    runHook preInstall
-
-    install -Dt "$out" kanjidic2.xml
-
-    runHook postInstall
-  '';
-
-  meta = edrdgMetadata // {
-    description = "A consolidated XML-format kanji database";
-    homepage = "https://www.edrdg.org/kanjidic/kanjd2index_legacy.html";
-  };
-}
--- a/nix/radkfile.nix
+++ b/nix/radkfile.nix
@@ -1,40 +0,0 @@
-{
-  stdenv,
-  radkfile-src,
-  gzip,
-  iconv,
-  edrdgMetadata,
-}:
-stdenv.mkDerivation {
-  name = "radkfile";
-
-  src = radkfile-src;
-  dontUnpack = true;
-
-  nativeBuildInputs = [
-    gzip
-    iconv
-  ];
-
-  buildPhase = ''
-    runHook preBuild
-
-    gzip -dkc "$src" > radkfile
-    iconv -f EUC-JP -t UTF-8 -o radkfile_utf8 radkfile
-
-    runHook postBuild
-  '';
-
-  installPhase = ''
-    runHook preInstall
-
-    install -Dt "$out" radkfile_utf8
-
-    runHook postInstall
-  '';
-
-  meta = edrdgMetadata // {
-    description = "A file providing searchable decompositions of kanji characters";
-    homepage = "https://www.edrdg.org/krad/kradinf.html";
-  };
-}
--- a/pubspec.lock
+++ b/pubspec.lock
@@ -5,18 +5,18 @@ packages:
    dependency: transitive
    description:
      name: _fe_analyzer_shared
-      sha256: "3b19a47f6ea7c2632760777c78174f47f6aec1e05f0cd611380d4593b8af1dbc"
+      sha256: "8d718c5c58904f9937290fd5dbf2d6a0e02456867706bfb6cd7b81d394e738d5"
      url: "https://pub.dev"
    source: hosted
-    version: "96.0.0"
+    version: "98.0.0"
  analyzer:
    dependency: transitive
    description:
      name: analyzer
-      sha256: "0c516bc4ad36a1a75759e54d5047cb9d15cded4459df01aa35a0b5ec7db2c2a0"
+      sha256: "6141ad5d092d1e1d13929c0504658bbeccc1703505830d7c26e859908f5efc88"
      url: "https://pub.dev"
    source: hosted
-    version: "10.2.0"
+    version: "12.0.0"
  args:
    dependency: "direct main"
    description:
@@ -29,10 +29,10 @@ packages:
    dependency: transitive
    description:
      name: async
-      sha256: "758e6d74e971c3e5aceb4110bfd6698efc7f501675bcfe0c775459a8140750eb"
+      sha256: e2eb0491ba5ddb6177742d2da23904574082139b07c1e33b8503b9f46f3e1a37
      url: "https://pub.dev"
    source: hosted
-    version: "2.13.0"
+    version: "2.13.1"
  benchmark_harness:
    dependency: "direct dev"
    description:
@@ -101,10 +101,10 @@ packages:
    dependency: "direct main"
    description:
      name: csv
-      sha256: bef2950f7a753eb82f894a2eabc3072e73cf21c17096296a5a992797e50b1d0d
+      sha256: "2e0a52fb729f2faacd19c9c0c954ff450bba37aa8ab999410309e2342e7013a2"
      url: "https://pub.dev"
    source: hosted
-    version: "7.1.0"
+    version: "8.0.0"
  equatable:
    dependency: "direct main"
    description:
@@ -149,10 +149,10 @@ packages:
    dependency: transitive
    description:
      name: hooks
-      sha256: "7a08a0d684cb3b8fb604b78455d5d352f502b68079f7b80b831c62220ab0a4f6"
+      sha256: e79ed1e8e1929bc6ecb6ec85f0cb519c887aa5b423705ded0d0f2d9226def388
      url: "https://pub.dev"
    source: hosted
-    version: "1.0.1"
+    version: "1.0.2"
  http_multi_server:
    dependency: transitive
    description:
@@ -197,18 +197,18 @@ packages:
    dependency: transitive
    description:
      name: matcher
-      sha256: "12956d0ad8390bbcc63ca2e1469c0619946ccb52809807067a7020d57e647aa6"
+      sha256: dc0b7dc7651697ea4ff3e69ef44b0407ea32c487a39fff6a4004fa585e901861
      url: "https://pub.dev"
    source: hosted
-    version: "0.12.18"
+    version: "0.12.19"
  meta:
    dependency: transitive
    description:
      name: meta
-      sha256: "9f29b9bcc8ee287b1a31e0d01be0eae99a930dbffdaecf04b3f3d82a969f296f"
+      sha256: df0c643f44ad098eb37988027a8e2b2b5a031fd3977f06bbfd3a76637e8df739
      url: "https://pub.dev"
    source: hosted
-    version: "1.18.1"
+    version: "1.18.2"
  mime:
    dependency: transitive
    description:
@@ -221,10 +221,10 @@ packages:
    dependency: transitive
    description:
      name: native_toolchain_c
-      sha256: "89e83885ba09da5fdf2cdacc8002a712ca238c28b7f717910b34bcd27b0d03ac"
+      sha256: "6ba77bb18063eebe9de401f5e6437e95e1438af0a87a3a39084fbd37c90df572"
      url: "https://pub.dev"
    source: hosted
-    version: "0.17.4"
+    version: "0.17.6"
  node_preamble:
    dependency: transitive
    description:
@@ -349,10 +349,10 @@ packages:
    dependency: "direct main"
    description:
      name: sqlite3
-      sha256: b7cf6b37667f6a921281797d2499ffc60fb878b161058d422064f0ddc78f6aa6
+      sha256: caa693ad15a587a2b4fde093b728131a1827903872171089dedb16f7665d3a91
      url: "https://pub.dev"
    source: hosted
-    version: "3.1.6"
+    version: "3.2.0"
  stack_trace:
    dependency: transitive
    description:
@@ -397,26 +397,26 @@ packages:
    dependency: "direct dev"
    description:
      name: test
-      sha256: "54c516bbb7cee2754d327ad4fca637f78abfc3cbcc5ace83b3eda117e42cd71a"
+      sha256: "8d9ceddbab833f180fbefed08afa76d7c03513dfdba87ffcec2718b02bbcbf20"
      url: "https://pub.dev"
    source: hosted
-    version: "1.29.0"
+    version: "1.31.0"
  test_api:
    dependency: transitive
    description:
      name: test_api
-      sha256: "93167629bfc610f71560ab9312acdda4959de4df6fac7492c89ff0d3886f6636"
+      sha256: "949a932224383300f01be9221c39180316445ecb8e7547f70a41a35bf421fb9e"
      url: "https://pub.dev"
    source: hosted
-    version: "0.7.9"
+    version: "0.7.11"
  test_core:
    dependency: transitive
    description:
      name: test_core
-      sha256: "394f07d21f0f2255ec9e3989f21e54d3c7dc0e6e9dbce160e5a9c1a6be0e2943"
+      sha256: "1991d4cfe85d5043241acac92962c3977c8d2f2add1ee73130c7b286417d1d34"
      url: "https://pub.dev"
    source: hosted
-    version: "0.6.15"
+    version: "0.6.17"
  typed_data:
    dependency: transitive
    description:
--- a/pubspec.yaml
+++ b/pubspec.yaml
@@ -9,7 +9,7 @@ environment:
 dependencies:
  args: ^2.7.0
  collection: ^1.19.0
-  csv: ^7.1.0
+  csv: ^8.0.0
  equatable: ^2.0.0
  path: ^1.9.1
  sqflite_common: ^2.5.0
--- a/test/search/radical_search_test.dart
+++ b/test/search/radical_search_test.dart
@@ -0,0 +1,79 @@
+import 'package:collection/collection.dart';
+import 'package:jadb/const_data/radicals.dart';
+import 'package:jadb/search.dart';
+import 'package:jadb/table_names/radkfile.dart';
+import 'package:test/test.dart';
+
+import 'setup_database_connection.dart';
+
+void main() {
+  test(
+    'All constant radicals should exist in the database',
+    () async {
+      final connection = await setupDatabaseConnection();
+      final allRadicalsInDb = await connection.query(
+        RADKFILETableNames.radkfile,
+        columns: ['radical'],
+        distinct: true,
+      );
+
+      final radicalsInDb = allRadicalsInDb
+          .map((e) => e['radical'] as String)
+          .toSet();
+
+      final missingRadicals = radicals.values.flattenedToSet.difference(
+        radicalsInDb,
+      );
+
+      expect(
+        missingRadicals,
+        isEmpty,
+        reason: 'Missing radicals in database: $missingRadicals',
+      );
+    },
+    skip: 'Test is valid, code is broken, fix me',
+  );
+
+  test(
+    'All radicals in database should be in the constant radical list',
+    () async {
+      final connection = await setupDatabaseConnection();
+      final allRadicalsInDb = await connection.query(
+        RADKFILETableNames.radkfile,
+        columns: ['radical'],
+        distinct: true,
+      );
+
+      final radicalsInDb = allRadicalsInDb
+          .map((e) => e['radical'] as String)
+          .toSet();
+
+      final extraRadicals = radicalsInDb.difference(
+        radicals.values.flattenedToSet,
+      );
+
+      expect(
+        extraRadicals,
+        isEmpty,
+        reason:
+            'Extra radicals in database missing in the constant list: $extraRadicals',
+      );
+    },
+    skip: 'Test is valid, code is broken, fix me',
+  );
+
+  group(
+    'All radicals should return results',
+    () {
+      for (final radical in radicals.values.flattened) {
+        test(' - $radical', () async {
+          final connection = await setupDatabaseConnection();
+          final result = await connection.jadbSearchKanjiByRadicals([radical]);
+          expect(result, isNotEmpty);
+        });
+      }
+    },
+    skip:
+        'These will be automatically fixed once the other radical tests are passing',
+  );
+}
Author	SHA1	Message	Date
h7x4	d168f07563	WIP All checks were successful Build and test / build (push) Successful in 17m29s Details	2026-04-13 21:10:56 +09:00
h7x4	d13138f8a5	Add datasource versions to database All checks were successful Build and test / build (push) Successful in 7m56s Details	2026-04-13 21:00:39 +09:00
h7x4	cbaa9ec6b3	benchmark: create separate benchmarks for jp and en search All checks were successful Build and test / build (push) Successful in 6m53s Details	2026-04-13 20:24:26 +09:00
h7x4	d1e2fa3748	test/search/radical_search: skip failing tests for now	2026-04-13 20:24:25 +09:00
h7x4	3f4fdf470d	jmdict: store glossary type in different table	2026-04-13 20:24:25 +09:00
h7x4	556d07913d	jmdict: don't store glossary language	2026-04-13 19:42:11 +09:00
h7x4	6165045ea7	migrations: simplify `JMdict_CombinedEntryScore`	2026-04-13 19:29:04 +09:00
h7x4	316dff3b46	migrations: comment out unused jmdict <-> kanjidic xref table	2026-04-13 19:27:16 +09:00
h7x4	747e680a02	migrations: remove some excessive indices	2026-04-13 19:27:16 +09:00
h7x4	4f73e07056	test/search/radical_search: init	2026-04-13 19:12:15 +09:00
h7x4	15540514f6	jmdict: don't store kanji + reading for xrefs All checks were successful Build and test / build (push) Successful in 7m0s Details	2026-04-13 18:33:06 +09:00
h7x4	4faf543d6e	jmdict: don't store empty entry scores All checks were successful Build and test / build (push) Successful in 8m17s Details	2026-04-13 18:18:48 +09:00
h7x4	d1a6f39cca	kanjidic: split `grade`/`freq`/`jlpt` into separate tables All checks were successful Build and test / build (push) Successful in 8m1s Details	2026-04-09 14:10:51 +09:00
h7x4	a222b2d9b8	jmdict: `elementId` instead of `reading` for element restriction tables All checks were successful Build and test / build (push) Successful in 8m0s Details	2026-04-08 19:57:07 +09:00
h7x4	6364457d9e	docs/database: add some notes about `elementId` embeddings All checks were successful Build and test / build (push) Successful in 7m56s Details	2026-04-08 19:07:48 +09:00
h7x4	5d26b41524	jmdict: embed element type (k/r) into `elementId`	2026-04-08 19:05:02 +09:00
h7x4	114febbe02	docs/database: add some notes about `JMdict_EntryScore` All checks were successful Build and test / build (push) Successful in 9m59s Details	2026-04-08 18:14:28 +09:00
h7x4	20243dec09	docs: add docs about database schema choices All checks were successful Build and test / build (push) Successful in 8m35s Details	2026-04-08 18:01:16 +09:00
h7x4	f6de8680ad	jmdict: infer `entryId` from element ids	2026-04-08 18:01:16 +09:00
h7x4	99218a6987	jmdict: embed `orderNum` in `senseId` for senses	2026-04-08 17:21:50 +09:00
h7x4	e8ee1ab944	data_ingestion/tanos-jlpt: remove redundant code import	2026-04-08 17:21:49 +09:00
h7x4	4f320e4ea9	jmdict: embed `orderNum` in `elementId` for kanji and readings	2026-04-08 17:21:49 +09:00
h7x4	9c9f5543c8	.gitea/workflows: fix 'update database inputs' step All checks were successful Build and test / build (push) Successful in 8m59s Details	2026-04-08 14:05:28 +09:00
h7x4	be493a6150	.gitea/workflows: fix build-and-test job All checks were successful Build and test / build (push) Successful in 8m59s Details	2026-04-08 13:36:42 +09:00
h7x4	8d742b92be	flake.nix: pull tanos jlpt data from `datasources` repo Some checks failed Build and test / build (push) Has been cancelled Details	2026-04-08 13:34:25 +09:00
h7x4	9b9c771eff	flake.nix: pull datasources from `datasources` repo Some checks failed Build and test / build (push) Failing after 8m19s Details	2026-04-07 17:26:47 +09:00
h7x4	eebeaba0e0	flake.nix: split off sqlite debugging tools into separate devshell Some checks failed Build and test / build (push) Failing after 11m9s Details	2026-04-06 12:56:17 +09:00
h7x4	61ac226fc3	word_search_result: add getter for unusual kanji flag All checks were successful Build and test / build (push) Successful in 11m48s Details	2026-04-02 15:53:39 +09:00
h7x4	ede57a7a00	docs: init All checks were successful Build and test / build (push) Successful in 11m51s Details	2026-04-01 16:48:40 +09:00
h7x4	2ad1e038f1	tanos-jlpt: remove `flatten` from xml stream All checks were successful Build and test / build (push) Successful in 13m41s Details This was earlier used to compensate for a double nesting bug. This has been fixed in the latest version of the xml package.	2026-04-01 16:04:44 +09:00
h7x4	f40825de65	jmdict: skip inserting duplicate xrefs	2026-04-01 16:03:56 +09:00
h7x4	5aa068eaec	flake.nix: add `sqldiff` to devshell Some checks failed Build and test / build (push) Failing after 12m0s Details	2026-04-01 15:27:25 +09:00
h7x4	170c3a853e	flake.lock: bump, pubspec.lock: update inputs Some checks failed Build and test / build (push) Failing after 10m4s Details	2026-03-26 22:18:10 +09:00