Oystein Kristoffer Tveit oysteikt
6 Followers · 22 Following
Overview Repositories
53
 Projects Packages Code Public Activity Starred Repositories
37
oysteikt pushed to main at mugiten/datasources 2026-04-13 13:53:37 +02:00
7fe3552bb1 flake.nix: add dates and hashes to all packages
oysteikt pushed to main at mugiten/jadb 2026-04-13 13:26:12 +02:00
cbaa9ec6b3 benchmark: create separate benchmarks for jp and en search
d1e2fa3748 test/search/radical_search: skip failing tests for now
3f4fdf470d jmdict: store glossary type in different table
Compare 3 commits »
oysteikt opened issue mugiten/jadb#103 2026-04-13 13:23:20 +02:00
Get rid of JMdict_SenseGlossary_byPhrase index for production use
oysteikt pushed to main at mugiten/jadb 2026-04-13 13:18:56 +02:00
34a30b3a35 benchmark: create separate benchmarks for jp and en search
oysteikt pushed to main at mugiten/jadb 2026-04-13 12:51:33 +02:00
0bb542840c test/search/radical_search: skip failing tests for now
oysteikt closed issue mugiten/jadb#100 2026-04-13 12:44:31 +02:00
Don't store JMdict_Sense.language when building with translations from a single language
oysteikt closed issue mugiten/jadb#101 2026-04-13 12:44:31 +02:00
Externalize JMdict_Sense.type into a separate table
oysteikt pushed to main at mugiten/jadb 2026-04-13 12:43:40 +02:00
507f1a21d8 jmdict: store glossary type in different table
556d07913d jmdict: don't store glossary language
6165045ea7 migrations: simplify JMdict_CombinedEntryScore
316dff3b46 migrations: comment out unused jmdict <-> kanjidic xref table
747e680a02 migrations: remove some excessive indices
Compare 5 commits »
oysteikt opened issue mugiten/jadb#102 2026-04-13 12:37:42 +02:00
Retrieve JMdict_SenseGlossaryType information
oysteikt commented on issue mugiten/jadb#90 2026-04-13 12:26:37 +02:00
Support wildcards in word search

Note that there are glossary phrases for * and _, see https://jisho.org/word/%E2%97%8B

oysteikt opened issue mugiten/jadb#101 2026-04-13 12:21:31 +02:00
Externalize JMdict_Sense.type into a separate table
oysteikt opened issue mugiten/jadb#100 2026-04-13 12:20:46 +02:00
Don't store JMdict_Sense.language when building with translations from a single language
oysteikt commented on issue mugiten/jadb#96 2026-04-13 12:15:08 +02:00
Use elementId for XREF__JMdict_KanjiElement__KANJIDIC_Character

This wasn't even used, commenting out the entire table for now

oysteikt closed issue mugiten/jadb#96 2026-04-13 12:15:08 +02:00
Use elementId for XREF__JMdict_KanjiElement__KANJIDIC_Character
oysteikt pushed to main at mugiten/jadb 2026-04-13 12:14:46 +02:00
c159ab3f4f migrations: comment out unused jmdict <-> kanjidic xref table
e89930a5ca migrations: remove some excessive indices
4f73e07056 test/search/radical_search: init
Compare 3 commits »
oysteikt pushed to main at mugiten/jadb 2026-04-13 11:33:22 +02:00
15540514f6 jmdict: don't store kanji + reading for xrefs
oysteikt closed issue mugiten/jadb#94 2026-04-13 11:33:14 +02:00
Don't store kanji + reading in xref tables
oysteikt closed issue mugiten/jadb#65 2026-04-13 11:26:10 +02:00
Omit rows with basescore 0 in JMdict_EntryScore
oysteikt commented on issue mugiten/jadb#65 2026-04-13 11:26:09 +02:00
Omit rows with basescore 0 in JMdict_EntryScore

Waow, we went from 162MB -> 137MB in db size and search benchmarks went from about 1400ms -> 920 ms

oysteikt pushed to main at mugiten/jadb 2026-04-13 11:19:04 +02:00
4faf543d6e jmdict: don't store empty entry scores