mirror of
https://github.com/mastodon/mastodon.git
synced 2024-12-23 09:35:03 +00:00
101 lines
3.1 KiB
Ruby
101 lines
3.1 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
class Importer::BaseImporter
|
|
# @param [Integer] batch_size
|
|
# @param [Concurrent::ThreadPoolExecutor] executor
|
|
def initialize(batch_size:, executor:)
|
|
@batch_size = batch_size
|
|
@executor = executor
|
|
@wait_for = Concurrent::Set.new
|
|
end
|
|
|
|
# Callback to run when a concurrent work unit completes
|
|
# @param [Proc]
|
|
def on_progress(&block)
|
|
@on_progress = block
|
|
end
|
|
|
|
# Callback to run when a concurrent work unit fails
|
|
# @param [Proc]
|
|
def on_failure(&block)
|
|
@on_failure = block
|
|
end
|
|
|
|
# Reduce resource usage during and improve speed of indexing
|
|
def optimize_for_import!
|
|
Chewy.client.indices.put_settings index: index.index_name, body: { index: { refresh_interval: -1 } }
|
|
end
|
|
|
|
# Restore original index settings
|
|
def optimize_for_search!
|
|
Chewy.client.indices.put_settings index: index.index_name, body: { index: { refresh_interval: index.settings_hash[:settings][:index][:refresh_interval] } }
|
|
end
|
|
|
|
# Estimate the amount of documents that would be indexed. Not exact!
|
|
# @returns [Integer]
|
|
def estimate!
|
|
reltuples = ActiveRecord::Base.connection_pool.with_connection { |connection| connection.select_one("SELECT reltuples FROM pg_class WHERE relname = '#{index.adapter.target.table_name}'")['reltuples'].to_i }
|
|
# If the table has never yet been vacuumed or analyzed, reltuples contains -1
|
|
[reltuples, 0].max
|
|
end
|
|
|
|
# Import data from the database into the index
|
|
def import!
|
|
raise NotImplementedError
|
|
end
|
|
|
|
# Remove documents from the index that no longer exist in the database
|
|
def clean_up!
|
|
index.scroll_batches do |documents|
|
|
primary_key = index.adapter.target.primary_key
|
|
raise ActiveRecord::UnknownPrimaryKey, index.adapter.target if primary_key.nil?
|
|
|
|
ids = documents.pluck('_id')
|
|
existence_map = index.adapter.target.where(primary_key => ids).pluck(primary_key).each_with_object({}) { |id, map| map[id.to_s] = true }
|
|
tmp = ids.reject { |id| existence_map[id] }
|
|
|
|
next if tmp.empty?
|
|
|
|
in_work_unit(tmp) do |deleted_ids|
|
|
bulk = Chewy::Index::Import::BulkBuilder.new(index, delete: deleted_ids).bulk_body
|
|
|
|
Chewy::Index::Import::BulkRequest.new(index).perform(bulk)
|
|
|
|
[0, bulk.size]
|
|
end
|
|
end
|
|
|
|
wait!
|
|
end
|
|
|
|
protected
|
|
|
|
def build_bulk_body(to_import)
|
|
# Specialize `Chewy::Index::Import::BulkBuilder#bulk_body` to avoid a few
|
|
# inefficiencies, as none of our fields or join fields and we do not need
|
|
# `BulkBuilder`'s versatility.
|
|
crutches = Chewy::Index::Crutch::Crutches.new index, to_import
|
|
to_import.map { |object| { index: { _id: object.id, data: index.compose(object, crutches, fields: []) } } }
|
|
end
|
|
|
|
def in_work_unit(...)
|
|
work_unit = Concurrent::Promises.future_on(@executor, ...)
|
|
|
|
work_unit.on_fulfillment!(&@on_progress)
|
|
work_unit.on_rejection!(&@on_failure)
|
|
work_unit.on_resolution! { @wait_for.delete(work_unit) }
|
|
|
|
@wait_for << work_unit
|
|
rescue Concurrent::RejectedExecutionError
|
|
sleep(0.1) && retry # Backpressure
|
|
end
|
|
|
|
def wait!
|
|
Concurrent::Promises.zip(*@wait_for).wait
|
|
end
|
|
|
|
def index
|
|
raise NotImplementedError
|
|
end
|
|
end
|