* Change: #532 ElasticSearch設定の外出し * バージョンチェック * 起動時にエラー
This commit is contained in:
KMY（雪あすか）
GitHub
parent
8e7c66522e
commit
a8fbcb3fb6
9 changed files with 508 additions and 269 deletions
3
.gitignore
vendored
3
.gitignore
vendored
|
|
@ -28,6 +28,9 @@
|
|||
/node_modules/
|
||||
/build/
|
||||
|
||||
# Ignore elasticsearch config
|
||||
/.elasticsearch.yml
|
||||
|
||||
# Ignore Vagrant files
|
||||
.vagrant/
|
||||
|
||||
|
|
|
|||
|
|
@ -3,83 +3,9 @@
|
|||
class AccountsIndex < Chewy::Index
|
||||
include DatetimeClampingConcern
|
||||
|
||||
settings index: index_preset(refresh_interval: '30s'), analysis: {
|
||||
filter: {
|
||||
english_stop: {
|
||||
type: 'stop',
|
||||
stopwords: '_english_',
|
||||
},
|
||||
|
||||
english_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'english',
|
||||
},
|
||||
|
||||
english_possessive_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'possessive_english',
|
||||
},
|
||||
|
||||
my_posfilter: {
|
||||
type: 'sudachi_part_of_speech',
|
||||
stoptags: [
|
||||
'助詞',
|
||||
'助動詞',
|
||||
'補助記号,句点',
|
||||
'補助記号,読点',
|
||||
],
|
||||
},
|
||||
},
|
||||
|
||||
analyzer: {
|
||||
natural: {
|
||||
tokenizer: 'standard',
|
||||
filter: %w(
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
elision
|
||||
english_possessive_stemmer
|
||||
english_stop
|
||||
english_stemmer
|
||||
),
|
||||
},
|
||||
|
||||
sudachi_analyzer: {
|
||||
filter: %w(
|
||||
my_posfilter
|
||||
sudachi_normalizedform
|
||||
),
|
||||
type: 'custom',
|
||||
tokenizer: 'sudachi_tokenizer',
|
||||
},
|
||||
|
||||
verbatim: {
|
||||
tokenizer: 'standard',
|
||||
filter: %w(lowercase asciifolding cjk_width),
|
||||
},
|
||||
|
||||
edge_ngram: {
|
||||
tokenizer: 'edge_ngram',
|
||||
filter: %w(lowercase asciifolding cjk_width),
|
||||
},
|
||||
},
|
||||
|
||||
tokenizer: {
|
||||
edge_ngram: {
|
||||
type: 'edge_ngram',
|
||||
min_gram: 1,
|
||||
max_gram: 15,
|
||||
},
|
||||
|
||||
sudachi_tokenizer: {
|
||||
resources_path: '/etc/elasticsearch/sudachi',
|
||||
split_mode: 'A',
|
||||
type: 'sudachi_tokenizer',
|
||||
discard_punctuation: 'true',
|
||||
},
|
||||
},
|
||||
}
|
||||
# ElasticSearch config is moved to "/config/elasticsearch.default.yml".
|
||||
# Edit it when original Mastodon changed ElasticSearch config.
|
||||
settings index: index_preset(refresh_interval: '30s'), analysis: ChewyConfig.instance.accounts
|
||||
|
||||
index_scope ::Account.searchable.includes(:account_stat)
|
||||
|
||||
|
|
@ -90,8 +16,15 @@ class AccountsIndex < Chewy::Index
|
|||
field(:properties, type: 'keyword', value: ->(account) { account.searchable_properties })
|
||||
field(:last_status_at, type: 'date', value: ->(account) { clamp_date(account.last_status_at || account.created_at) })
|
||||
field(:domain, type: 'keyword', value: ->(account) { account.domain || '' })
|
||||
field(:display_name, type: 'text', analyzer: 'verbatim') { field :edge_ngram, type: 'text', analyzer: 'edge_ngram', search_analyzer: 'verbatim' }
|
||||
field(:username, type: 'text', analyzer: 'verbatim', value: ->(account) { [account.username, account.domain].compact.join('@') }) { field :edge_ngram, type: 'text', analyzer: 'edge_ngram', search_analyzer: 'verbatim' }
|
||||
field(:text, type: 'text', analyzer: 'sudachi_analyzer', value: ->(account) { account.searchable_text }) { field(:stemmed, type: 'text', analyzer: 'natural') }
|
||||
field(:display_name, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('display_name', 'analyzer')) do
|
||||
field :edge_ngram, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('display_name', 'edge_ngram', 'analyzer'), search_analyzer: ChewyConfig.instance.accounts_analyzers.dig('display_name', 'edge_ngram', 'search_analyzer')
|
||||
end
|
||||
field(:username, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('username', 'analyzer'), value: lambda { |account|
|
||||
[account.username, account.domain].compact.join('@')
|
||||
}) do
|
||||
field :edge_ngram, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('username', 'edge_ngram', 'analyzer'),
|
||||
search_analyzer: ChewyConfig.instance.accounts_analyzers.dig('username', 'edge_ngram', 'search_analyzer')
|
||||
end
|
||||
field(:text, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('text', 'analyzer'), value: ->(account) { account.searchable_text }) { field(:stemmed, type: 'text', analyzer: ChewyConfig.instance.accounts_analyzers.dig('text', 'stemmed', 'analyzer')) }
|
||||
end
|
||||
end
|
||||
|
|
|
|||
|
|
@ -3,81 +3,9 @@
|
|||
class PublicStatusesIndex < Chewy::Index
|
||||
include DatetimeClampingConcern
|
||||
|
||||
settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: {
|
||||
filter: {
|
||||
english_stop: {
|
||||
type: 'stop',
|
||||
stopwords: '_english_',
|
||||
},
|
||||
|
||||
english_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'english',
|
||||
},
|
||||
|
||||
english_possessive_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'possessive_english',
|
||||
},
|
||||
|
||||
my_posfilter: {
|
||||
type: 'sudachi_part_of_speech',
|
||||
stoptags: [
|
||||
'助詞',
|
||||
'助動詞',
|
||||
'補助記号,句点',
|
||||
'補助記号,読点',
|
||||
],
|
||||
},
|
||||
},
|
||||
|
||||
analyzer: {
|
||||
content: {
|
||||
tokenizer: 'uax_url_email',
|
||||
filter: %w(
|
||||
english_possessive_stemmer
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
english_stop
|
||||
english_stemmer
|
||||
),
|
||||
},
|
||||
|
||||
hashtag: {
|
||||
tokenizer: 'keyword',
|
||||
filter: %w(
|
||||
word_delimiter_graph
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
),
|
||||
},
|
||||
|
||||
sudachi_analyzer: {
|
||||
tokenizer: 'sudachi_tokenizer',
|
||||
type: 'custom',
|
||||
filter: %w(
|
||||
english_possessive_stemmer
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
english_stop
|
||||
english_stemmer
|
||||
my_posfilter
|
||||
sudachi_normalizedform
|
||||
),
|
||||
},
|
||||
},
|
||||
tokenizer: {
|
||||
sudachi_tokenizer: {
|
||||
resources_path: '/etc/elasticsearch/sudachi',
|
||||
split_mode: 'A',
|
||||
type: 'sudachi_tokenizer',
|
||||
discard_punctuation: 'true',
|
||||
},
|
||||
},
|
||||
}
|
||||
# ElasticSearch config is moved to "/config/elasticsearch.default.yml".
|
||||
# Edit it when original Mastodon changed ElasticSearch config.
|
||||
settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: ChewyConfig.instance.public_statuses
|
||||
|
||||
index_scope ::Status.unscoped
|
||||
.kept
|
||||
|
|
@ -87,8 +15,8 @@ class PublicStatusesIndex < Chewy::Index
|
|||
root date_detection: false do
|
||||
field(:id, type: 'long')
|
||||
field(:account_id, type: 'long')
|
||||
field(:text, type: 'text', analyzer: 'sudachi_analyzer', value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: 'content') }
|
||||
field(:tags, type: 'text', analyzer: 'hashtag', value: ->(status) { status.tags.map(&:display_name) })
|
||||
field(:text, type: 'text', analyzer: ChewyConfig.instance.public_statuses_analyzers.dig('text', 'analyzer'), value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: ChewyConfig.instance.public_statuses_analyzers.dig('text', 'stemmed', 'analyzer')) }
|
||||
field(:tags, type: 'text', analyzer: ChewyConfig.instance.public_statuses_analyzers.dig('tags', 'analyzer'), value: ->(status) { status.tags.map(&:display_name) })
|
||||
field(:language, type: 'keyword')
|
||||
field(:domain, type: 'keyword', value: ->(status) { status.account.domain || '' })
|
||||
field(:properties, type: 'keyword', value: ->(status) { status.searchable_properties })
|
||||
|
|
|
|||
|
|
@ -3,85 +3,9 @@
|
|||
class StatusesIndex < Chewy::Index
|
||||
include DatetimeClampingConcern
|
||||
|
||||
settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: {
|
||||
filter: {
|
||||
english_stop: {
|
||||
type: 'stop',
|
||||
stopwords: '_english_',
|
||||
},
|
||||
|
||||
english_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'english',
|
||||
},
|
||||
|
||||
english_possessive_stemmer: {
|
||||
type: 'stemmer',
|
||||
language: 'possessive_english',
|
||||
},
|
||||
|
||||
my_posfilter: {
|
||||
type: 'sudachi_part_of_speech',
|
||||
stoptags: [
|
||||
'助詞',
|
||||
'助動詞',
|
||||
'補助記号,句点',
|
||||
'補助記号,読点',
|
||||
],
|
||||
},
|
||||
},
|
||||
analyzer: {
|
||||
verbatim: {
|
||||
tokenizer: 'uax_url_email',
|
||||
filter: %w(lowercase),
|
||||
},
|
||||
|
||||
content: {
|
||||
tokenizer: 'uax_url_email',
|
||||
filter: %w(
|
||||
english_possessive_stemmer
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
english_stop
|
||||
english_stemmer
|
||||
),
|
||||
},
|
||||
|
||||
hashtag: {
|
||||
tokenizer: 'keyword',
|
||||
filter: %w(
|
||||
word_delimiter_graph
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
),
|
||||
},
|
||||
|
||||
sudachi_analyzer: {
|
||||
tokenizer: 'sudachi_tokenizer',
|
||||
type: 'custom',
|
||||
filter: %w(
|
||||
english_possessive_stemmer
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
english_stop
|
||||
english_stemmer
|
||||
my_posfilter
|
||||
sudachi_normalizedform
|
||||
),
|
||||
},
|
||||
},
|
||||
tokenizer: {
|
||||
sudachi_tokenizer: {
|
||||
resources_path: '/etc/elasticsearch/sudachi',
|
||||
split_mode: 'A',
|
||||
type: 'sudachi_tokenizer',
|
||||
discard_punctuation: 'true',
|
||||
},
|
||||
},
|
||||
}
|
||||
# ElasticSearch config is moved to "/config/elasticsearch.default.yml".
|
||||
# Edit it when original Mastodon changed ElasticSearch config.
|
||||
settings index: index_preset(refresh_interval: '30s', number_of_shards: 5), analysis: ChewyConfig.instance.statuses
|
||||
|
||||
index_scope ::Status.unscoped.kept.without_reblogs.includes(
|
||||
:account,
|
||||
|
|
@ -107,8 +31,8 @@ class StatusesIndex < Chewy::Index
|
|||
root date_detection: false do
|
||||
field(:id, type: 'long')
|
||||
field(:account_id, type: 'long')
|
||||
field(:text, type: 'text', analyzer: 'sudachi_analyzer', value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: 'content') }
|
||||
field(:tags, type: 'text', analyzer: 'hashtag', value: ->(status) { status.tags.map(&:display_name) })
|
||||
field(:text, type: 'text', analyzer: ChewyConfig.instance.statuses_analyzers.dig('text', 'analyzer'), value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: ChewyConfig.instance.statuses_analyzers.dig('text', 'stemmed', 'analyzer')) }
|
||||
field(:tags, type: 'text', analyzer: ChewyConfig.instance.statuses_analyzers.dig('tags', 'analyzer'), value: ->(status) { status.tags.map(&:display_name) })
|
||||
field(:searchable_by, type: 'long', value: ->(status) { status.searchable_by })
|
||||
field(:mentioned_by, type: 'long', value: ->(status) { status.mentioned_by })
|
||||
field(:favourited_by, type: 'long', value: ->(status) { status.favourited_by })
|
||||
|
|
|
|||
|
|
@ -3,36 +3,9 @@
|
|||
class TagsIndex < Chewy::Index
|
||||
include DatetimeClampingConcern
|
||||
|
||||
settings index: index_preset(refresh_interval: '30s'), analysis: {
|
||||
analyzer: {
|
||||
content: {
|
||||
tokenizer: 'keyword',
|
||||
filter: %w(
|
||||
word_delimiter_graph
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
),
|
||||
},
|
||||
|
||||
edge_ngram: {
|
||||
tokenizer: 'edge_ngram',
|
||||
filter: %w(
|
||||
lowercase
|
||||
asciifolding
|
||||
cjk_width
|
||||
),
|
||||
},
|
||||
},
|
||||
|
||||
tokenizer: {
|
||||
edge_ngram: {
|
||||
type: 'edge_ngram',
|
||||
min_gram: 2,
|
||||
max_gram: 15,
|
||||
},
|
||||
},
|
||||
}
|
||||
# ElasticSearch config is moved to "/config/elasticsearch.default.yml".
|
||||
# Edit it when original Mastodon changed ElasticSearch config.
|
||||
settings index: index_preset(refresh_interval: '30s'), analysis: ChewyConfig.instance.tags
|
||||
|
||||
index_scope ::Tag.listable
|
||||
|
||||
|
|
@ -41,7 +14,9 @@ class TagsIndex < Chewy::Index
|
|||
end
|
||||
|
||||
root date_detection: false do
|
||||
field(:name, type: 'text', analyzer: 'content', value: :display_name) { field(:edge_ngram, type: 'text', analyzer: 'edge_ngram', search_analyzer: 'content') }
|
||||
field(:name, type: 'text', analyzer: ChewyConfig.instance.tags_analyzers.dig('name', 'analyzer'), value: :display_name) do
|
||||
field(:edge_ngram, type: 'text', analyzer: ChewyConfig.instance.tags_analyzers.dig('name', 'edge_ngram', 'analyzer'), search_analyzer: ChewyConfig.instance.tags_analyzers.dig('name', 'edge_ngram', 'search_analyzer'))
|
||||
end
|
||||
field(:reviewed, type: 'boolean', value: ->(tag) { tag.reviewed? })
|
||||
field(:usage, type: 'long', value: ->(tag, crutches) { tag.history.aggregate(crutches.time_period).accounts })
|
||||
field(:last_status_at, type: 'date', value: ->(tag) { clamp_date(tag.last_status_at || tag.created_at) })
|
||||
|
|
|
|||
60
app/lib/chewy_config.rb
Normal file
60
app/lib/chewy_config.rb
Normal file
|
|
@ -0,0 +1,60 @@
|
|||
# frozen_string_literal: true
|
||||
|
||||
require 'singleton'
|
||||
require 'yaml'
|
||||
|
||||
class ChewyConfig
|
||||
include Singleton
|
||||
|
||||
class InvalidElasticSearchVersionError < Mastodon::Error; end
|
||||
|
||||
CONFIG_VERSION = 1
|
||||
|
||||
def initialize
|
||||
custom_config_file = Rails.root.join('.elasticsearch.yml')
|
||||
default_config_file = Rails.root.join('config', 'elasticsearch.default.yml')
|
||||
|
||||
custom_config = nil
|
||||
custom_config = YAML.load_file(custom_config_file) if File.exist?(custom_config_file)
|
||||
default_config = YAML.load_file(default_config_file)
|
||||
|
||||
@config = default_config.merge(custom_config || {})
|
||||
@config = @config.merge(YAML.load_file(Rails.root.join('config', 'elasticsearch.default-ja-sudachi.yml'))) if Rails.env.test?
|
||||
|
||||
raise InvalidElasticSearchVersionError, "ElasticSearch config version is missmatch. expected version=#{CONFIG_VERSION} actual version=#{@config['version']}" if @config['version'] != CONFIG_VERSION
|
||||
end
|
||||
|
||||
attr_reader :config
|
||||
|
||||
def accounts
|
||||
config['accounts']
|
||||
end
|
||||
|
||||
def accounts_analyzers
|
||||
config['accounts_analyzers']
|
||||
end
|
||||
|
||||
def public_statuses
|
||||
config['public_statuses']
|
||||
end
|
||||
|
||||
def public_statuses_analyzers
|
||||
config['public_statuses_analyzers']
|
||||
end
|
||||
|
||||
def statuses
|
||||
config['statuses']
|
||||
end
|
||||
|
||||
def statuses_analyzers
|
||||
config['statuses_analyzers']
|
||||
end
|
||||
|
||||
def tags
|
||||
config['tags']
|
||||
end
|
||||
|
||||
def tags_analyzers
|
||||
config['tags_analyzers']
|
||||
end
|
||||
end
|
||||
234
config/elasticsearch.default-ja-sudachi.yml
Normal file
234
config/elasticsearch.default-ja-sudachi.yml
Normal file
|
|
@ -0,0 +1,234 @@
|
|||
# This is a configuration file for environments that use Japanese and Sudachi plug-ins.
|
||||
# To use this file, copy it to the Mastodon root directory and rename the file to ".elasticsearch.yml".
|
||||
|
||||
version: 1
|
||||
|
||||
accounts:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
my_posfilter:
|
||||
type: sudachi_part_of_speech
|
||||
stoptags:
|
||||
- 助詞
|
||||
- 助動詞
|
||||
- 補助記号,句点
|
||||
- 補助記号,読点
|
||||
|
||||
analyzer:
|
||||
natural:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- elision
|
||||
- english_possessive_stemmer
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
sudachi_analyzer:
|
||||
type: custom
|
||||
tokenizer: sudachi_tokenizer
|
||||
filter:
|
||||
- my_posfilter
|
||||
- sudachi_normalizedform
|
||||
verbatim:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
edge_ngram:
|
||||
tokenizer: edge_ngram
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
tokenizer:
|
||||
edge_ngram:
|
||||
type: edge_ngram
|
||||
min_gram: 1
|
||||
max_gram: 15
|
||||
sudachi_tokenizer:
|
||||
resources_path: '/etc/elasticsearch/sudachi'
|
||||
split_mode: A
|
||||
type: sudachi_tokenizer
|
||||
discard_punctuation: 'true'
|
||||
|
||||
public_statuses:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
my_posfilter:
|
||||
type: sudachi_part_of_speech
|
||||
stoptags:
|
||||
- 助詞
|
||||
- 助動詞
|
||||
- 補助記号,句点
|
||||
- 補助記号,読点
|
||||
|
||||
analyzer:
|
||||
content:
|
||||
tokenizer: uax_url_email
|
||||
filter:
|
||||
- english_possessive_stemmer
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
hashtag:
|
||||
tokenizer: keyword
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
sudachi_analyzer:
|
||||
tokenizer: sudachi_tokenizer
|
||||
type: custom
|
||||
filter:
|
||||
- english_possessive_stemmer
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
- my_posfilter
|
||||
- sudachi_normalizedform
|
||||
|
||||
tokenizer:
|
||||
sudachi_tokenizer:
|
||||
resources_path: '/etc/elasticsearch/sudachi'
|
||||
split_mode: A
|
||||
type: sudachi_tokenizer
|
||||
discard_punctuation: 'true'
|
||||
|
||||
statuses:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
my_posfilter:
|
||||
type: sudachi_part_of_speech
|
||||
stoptags:
|
||||
- 助詞
|
||||
- 助動詞
|
||||
- 補助記号,句点
|
||||
- 補助記号,読点
|
||||
|
||||
analyzer:
|
||||
verbatim:
|
||||
tokenizer: uax_url_email
|
||||
filter:
|
||||
- lowercase
|
||||
content:
|
||||
tokenizer: uax_url_email
|
||||
filter:
|
||||
- english_possessive_stemmer
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
hashtag:
|
||||
tokenizer: keyword
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
sudachi_analyzer:
|
||||
tokenizer: sudachi_tokenizer
|
||||
type: custom
|
||||
filter:
|
||||
- english_possessive_stemmer
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
- my_posfilter
|
||||
- sudachi_normalizedform
|
||||
|
||||
tags:
|
||||
analyzer:
|
||||
content:
|
||||
tokenizer: keyword
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
edge_ngram:
|
||||
tokenizer: edge_ngram
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
tokenizer:
|
||||
edge_ngram:
|
||||
type: edge_ngram
|
||||
min_gram: 2
|
||||
max_gram: 15
|
||||
|
||||
accounts_analyzers:
|
||||
display_name:
|
||||
analyzer: verbatim
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: verbatim
|
||||
username:
|
||||
analyzer: verbatim
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: verbatim
|
||||
text:
|
||||
analyzer: sudachi_analyzer
|
||||
stemmed:
|
||||
analyzer: natural
|
||||
|
||||
public_statuses_analyzers:
|
||||
text:
|
||||
analyzer: sudachi_analyzer
|
||||
stemmed:
|
||||
analyzer: content
|
||||
tags:
|
||||
analyzer: hashtag
|
||||
|
||||
statuses_analyzers:
|
||||
text:
|
||||
analyzer: sudachi_analyzer
|
||||
stemmed:
|
||||
analyzer: content
|
||||
tags:
|
||||
analyzer: hashtag
|
||||
|
||||
tags_analyzers:
|
||||
name:
|
||||
analyzer: content
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: content
|
||||
177
config/elasticsearch.default.yml
Normal file
177
config/elasticsearch.default.yml
Normal file
|
|
@ -0,0 +1,177 @@
|
|||
# The standard ElasticSearch settings described in the original Mastodon code are stored.
|
||||
# This configuration file is overridden by creating a ".elasticsearch.yml" file in the Mastodon root directory.
|
||||
|
||||
version: 1
|
||||
|
||||
accounts:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
|
||||
analyzer:
|
||||
natural:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- elision
|
||||
- english_possessive_stemmer
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
verbatim:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
edge_ngram:
|
||||
tokenizer: edge_ngram
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
tokenizer:
|
||||
edge_ngram:
|
||||
type: edge_ngram
|
||||
min_gram: 1
|
||||
max_gram: 15
|
||||
|
||||
public_statuses:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
|
||||
analyzer:
|
||||
verbatim:
|
||||
tokenizer: uax_url_email
|
||||
filter:
|
||||
- lowercase
|
||||
content:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- elision
|
||||
- english_possessive_stemmer
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
hashtag:
|
||||
tokenizer: keyword
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
statuses:
|
||||
filter:
|
||||
english_stop:
|
||||
type: stop
|
||||
stopwords: _english_
|
||||
english_stemmer:
|
||||
type: stemmer
|
||||
language: english
|
||||
english_possessive_stemmer:
|
||||
type: stemmer
|
||||
language: possessive_english
|
||||
|
||||
analyzer:
|
||||
verbatim:
|
||||
tokenizer: uax_url_email
|
||||
filter:
|
||||
- lowercase
|
||||
content:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
- elision
|
||||
- english_possessive_stemmer
|
||||
- english_stop
|
||||
- english_stemmer
|
||||
hashtag:
|
||||
tokenizer: standard
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
tags:
|
||||
analyzer:
|
||||
content:
|
||||
tokenizer: keyword
|
||||
filter:
|
||||
- word_delimiter_graph
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
edge_ngram:
|
||||
tokenizer: edge_ngram
|
||||
filter:
|
||||
- lowercase
|
||||
- asciifolding
|
||||
- cjk_width
|
||||
|
||||
tokenizer:
|
||||
edge_ngram:
|
||||
type: edge_ngram
|
||||
min_gram: 2
|
||||
max_gram: 15
|
||||
|
||||
accounts_analyzers:
|
||||
display_name:
|
||||
analyzer: verbatim
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: verbatim
|
||||
username:
|
||||
analyzer: verbatim
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: verbatim
|
||||
text:
|
||||
analyzer: verbatim
|
||||
stemmed:
|
||||
analyzer: natural
|
||||
|
||||
public_statuses_analyzers:
|
||||
text:
|
||||
analyzer: verbatim
|
||||
stemmed:
|
||||
analyzer: content
|
||||
tags:
|
||||
analyzer: hashtag
|
||||
|
||||
statuses_analyzers:
|
||||
text:
|
||||
analyzer: verbatim
|
||||
stemmed:
|
||||
analyzer: content
|
||||
tags:
|
||||
analyzer: hashtag
|
||||
|
||||
tags_analyzers:
|
||||
name:
|
||||
analyzer: content
|
||||
edge_ngram:
|
||||
analyzer: edge_ngram
|
||||
search_analyzer: content
|
||||
|
|
@ -1,5 +1,7 @@
|
|||
# frozen_string_literal: true
|
||||
|
||||
require_relative '../../app/lib/chewy_config'
|
||||
|
||||
enabled = ENV['ES_ENABLED'] == 'true'
|
||||
host = ENV.fetch('ES_HOST') { 'localhost' }
|
||||
port = ENV.fetch('ES_PORT') { 9200 }
|
||||
|
|
@ -37,3 +39,6 @@ Chewy.use_after_commit_callbacks = false
|
|||
# Mastodon is run with hidden services enabled, because
|
||||
# Elasticsearch is *not* supposed to be accessed through a proxy
|
||||
Faraday.ignore_env_proxy = true
|
||||
|
||||
# Check Elasticsearch config file version
|
||||
ChewyConfig.instance.accounts if enabled
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue