From 41eeb9ebaa65abe3fcbab60847b69d2469726d8a Mon Sep 17 00:00:00 2001 From: Akihiko Odaki Date: Tue, 25 Aug 2020 20:39:35 +0900 Subject: [PATCH] Use Status.group instead of Status.distinct in HashQueryService (#14662) DISTINCT clause removes duplicated records according to all the selected attributes. In reality, it can remove duplicated records only looking at statuses.id, but the clause confuses the query planner and yields insufficient performance. The behavior is also problematic if the scope produced by HashQueryService is used to query columns without id (using pluck method, for example). The scope is expected to contain unique statuses, but the uniquness will be evaluated with some arbitrary columns other than id. GROUP BY clause resolves those problem by explicitly specifying the column to take into account for the record distinction. A workaround for the problem of DISTINCT clause in Api::V1::Timelines::TagController is no longer necessary and removed. --- app/controllers/api/v1/timelines/tag_controller.rb | 4 +--- app/services/hashtag_query_service.rb | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/app/controllers/api/v1/timelines/tag_controller.rb b/app/controllers/api/v1/timelines/tag_controller.rb index 2d6ad5a80..62f34d3f7 100644 --- a/app/controllers/api/v1/timelines/tag_controller.rb +++ b/app/controllers/api/v1/timelines/tag_controller.rb @@ -33,9 +33,7 @@ class Api::V1::Timelines::TagController < Api::BaseController ) if truthy_param?(:only_media) - # `SELECT DISTINCT id, updated_at` is too slow, so pluck ids at first, and then select id, updated_at with ids. - status_ids = statuses.joins(:media_attachments).distinct(:id).pluck(:id) - statuses.where(id: status_ids) + statuses.joins(:media_attachments) else statuses end diff --git a/app/services/hashtag_query_service.rb b/app/services/hashtag_query_service.rb index 196de0639..0bdf60221 100644 --- a/app/services/hashtag_query_service.rb +++ b/app/services/hashtag_query_service.rb @@ -8,7 +8,7 @@ class HashtagQueryService < BaseService all = tags_for(params[:all]) none = tags_for(params[:none]) - Status.distinct + Status.group(:id) .as_tag_timeline(tags, account, local) .tagged_with_all(all) .tagged_with_none(none)