Add new public status index (#26344)

Co-authored-by: Eugen Rochko <eugen@zeonfederated.com>
Co-authored-by: Claire <claire.github-309c@sitedethib.com>
This commit is contained in:
jsgoldstein 2023-08-24 10:40:04 -04:00 committed by GitHub
commit 30c191aaa0
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
28 changed files with 584 additions and 87 deletions

View file

@ -62,6 +62,6 @@ class AccountsIndex < Chewy::Index
field(:last_status_at, type: 'date', value: ->(account) { account.last_status_at || account.created_at })
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', value: ->(account) { account.searchable_text }) { field :stemmed, type: 'text', analyzer: 'natural' }
field(:text, type: 'text', analyzer: 'whitespace', value: ->(account) { account.searchable_text }) { field :stemmed, type: 'text', analyzer: 'natural' }
end
end

View file

@ -0,0 +1,50 @@
# frozen_string_literal: true
class PublicStatusesIndex < Chewy::Index
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',
},
},
analyzer: {
content: {
tokenizer: 'uax_url_email',
filter: %w(
english_possessive_stemmer
lowercase
asciifolding
cjk_width
english_stop
english_stemmer
),
},
},
}
index_scope ::Status.unscoped
.kept
.indexable
.includes(:media_attachments, :preloadable_poll, :preview_cards)
root date_detection: false do
field(:id, type: 'keyword')
field(:account_id, type: 'long')
field(:text, type: 'text', analyzer: 'whitespace', value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: 'content') }
field(:language, type: 'keyword')
field(:properties, type: 'keyword', value: ->(status) { status.searchable_properties })
field(:created_at, type: 'date')
end
end

View file

@ -1,23 +1,24 @@
# frozen_string_literal: true
class StatusesIndex < Chewy::Index
include FormattingHelper
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',
},
},
analyzer: {
content: {
tokenizer: 'uax_url_email',
@ -35,7 +36,7 @@ class StatusesIndex < Chewy::Index
# We do not use delete_if option here because it would call a method that we
# expect to be called with crutches without crutches, causing n+1 queries
index_scope ::Status.unscoped.kept.without_reblogs.includes(:media_attachments, :preloadable_poll)
index_scope ::Status.unscoped.kept.without_reblogs.includes(:media_attachments, :preloadable_poll, :preview_cards)
crutch :mentions do |collection|
data = ::Mention.where(status_id: collection.map(&:id)).where(account: Account.local, silent: false).pluck(:status_id, :account_id)
@ -63,13 +64,12 @@ class StatusesIndex < Chewy::Index
end
root date_detection: false do
field :id, type: 'long'
field :account_id, type: 'long'
field :text, type: 'text', value: ->(status) { status.searchable_text } do
field :stemmed, type: 'text', analyzer: 'content'
end
field :searchable_by, type: 'long', value: ->(status, crutches) { status.searchable_by(crutches) }
field(:id, type: 'keyword')
field(:account_id, type: 'long')
field(:text, type: 'text', analyzer: 'whitespace', value: ->(status) { status.searchable_text }) { field(:stemmed, type: 'text', analyzer: 'content') }
field(:searchable_by, type: 'long', value: ->(status, crutches) { status.searchable_by(crutches) })
field(:language, type: 'keyword')
field(:properties, type: 'keyword', value: ->(status) { status.searchable_properties })
field(:created_at, type: 'date')
end
end