Compare commits

...

2 Commits

Author SHA1 Message Date
758
64594aa0c8 CHANGELOGを更新(7p3) 2024-10-07 22:34:30 +09:00
758
978286baa5 kuromoji向けにカスタマイズ 2024-10-07 21:50:45 +09:00
4 changed files with 58 additions and 11 deletions

View File

@ -2,6 +2,11 @@
死闘の記録です
## [7p3] - 2024-10-07
Elasticsearchの設定 `/app/chewy/` を日本語検索(kuromoji, ICU analysis)向けに最適化
うまく動作しているかは分かりません
## [7p2] - 2024-09-08
3つのカスタムテーマを追加

View File

@ -23,13 +23,22 @@ class AccountsIndex < Chewy::Index
analyzer: {
natural: {
tokenizer: 'standard',
tokenizer: 'kuromoji',
type: 'custom',
char_filter: %w(
icu_normalizer
html_strip
kuromoji_iteration_mark
),
filter: %w(
english_possessive_stemmer
lowercase
asciifolding
kuromoji_stemmer
kuromoji_number
kuromoji_baseform
icu_normalizer
cjk_width
elision
english_possessive_stemmer
english_stop
english_stemmer
),
@ -52,6 +61,10 @@ class AccountsIndex < Chewy::Index
min_gram: 1,
max_gram: 15,
},
kuromoji: {
type: 'kuromoji_tokenizer',
mode: 'search',
},
},
}

View File

@ -20,7 +20,12 @@ class PublicStatusesIndex < Chewy::Index
language: 'possessive_english',
},
},
tokenizer: {
kuromoji: {
type: 'kuromoji_tokenizer',
mode: 'search',
},
},
analyzer: {
verbatim: {
tokenizer: 'uax_url_email',
@ -28,13 +33,22 @@ class PublicStatusesIndex < Chewy::Index
},
content: {
tokenizer: 'standard',
tokenizer: 'kuromoji',
type: 'custom',
char_filter: %w(
icu_normalizer
html_strip
kuromoji_iteration_mark
),
filter: %w(
english_possessive_stemmer
lowercase
asciifolding
kuromoji_stemmer
kuromoji_number
kuromoji_baseform
icu_normalizer
cjk_width
elision
english_possessive_stemmer
english_stop
english_stemmer
),

View File

@ -1,6 +1,7 @@
# frozen_string_literal: true
class StatusesIndex < Chewy::Index
include FormattingHelper
include DatetimeClampingConcern
settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: {
@ -20,7 +21,12 @@ class StatusesIndex < Chewy::Index
language: 'possessive_english',
},
},
tokenizer: {
kuromoji: {
type: 'kuromoji_tokenizer',
mode: 'search',
},
},
analyzer: {
verbatim: {
tokenizer: 'uax_url_email',
@ -28,13 +34,22 @@ class StatusesIndex < Chewy::Index
},
content: {
tokenizer: 'standard',
tokenizer: 'kuromoji',
type: 'custom',
char_filter: %w(
icu_normalizer
html_strip
kuromoji_iteration_mark
),
filter: %w(
english_possessive_stemmer
lowercase
asciifolding
kuromoji_stemmer
kuromoji_number
kuromoji_baseform
icu_normalizer
cjk_width
elision
english_possessive_stemmer
english_stop
english_stemmer
),