Precalculate trending projectsprecalculate-trending-projects

This commit introduces a Sidekiq worker that precalculates the list of
trending projects on a daily basis. The resulting set is stored in a
database table that is then queried by Project.trending.

This setup means that Unicorn workers no longer _may_ have to calculate
the list of trending projects. Furthermore it supports filtering without
any complex caching mechanisms.

The data in the "trending_projects" table is inserted in the same order
as the project ranking. This means that getting the projects in the
correct order is simply a matter of:

    SELECT projects.*
    FROM projects
    INNER JOIN trending_projects ON trending_projects.project_id = projects.id
    ORDER BY trending_projects.id ASC;

Such a query will only take a few milliseconds at most (as measured on
GitLab.com), opposed to a few seconds for the query used for calculating
the project ranks.

The migration in this commit does not require downtime and takes care of
populating an initial list of trending projects.

4 files changed, 71 insertions, 70 deletions

diff --git a/spec/finders/trending_projects_finder_spec.rb b/spec/finders/trending_projects_finder_spec.rb
deleted file mode 100644
index cfe15b9defa..00000000000
--- a/spec/finders/trending_projects_finder_spec.rb
+++ /dev/null
@@ -1,48 +0,0 @@
-require 'spec_helper'
-
-describe TrendingProjectsFinder do
-  let(:user) { create(:user) }
-  let(:public_project1) { create(:empty_project, :public) }
-  let(:public_project2) { create(:empty_project, :public) }
-  let(:private_project) { create(:empty_project, :private) }
-  let(:internal_project) { create(:empty_project, :internal) }
-
-  before do
-    3.times do
-      create(:note_on_commit, project: public_project1)
-    end
-
-    2.times do
-      create(:note_on_commit, project: public_project2, created_at: 5.weeks.ago)
-    end
-
-    create(:note_on_commit, project: private_project)
-    create(:note_on_commit, project: internal_project)
-  end
-
-  describe '#execute', caching: true do
-    context 'without an explicit time range' do
-      it 'returns public trending projects' do
-        projects = described_class.new.execute
-
-        expect(projects).to eq([public_project1])
-      end
-    end
-
-    context 'with an explicit time range' do
-      it 'returns public trending projects' do
-        projects = described_class.new.execute(2)
-
-        expect(projects).to eq([public_project1, public_project2])
-      end
-    end
-
-    it 'caches the list of projects' do
-      projects = described_class.new
-
-      expect(Project).to receive(:trending).once
-
-      2.times { projects.execute }
-    end
-  end
-end
diff --git a/spec/models/project_spec.rb b/spec/models/project_spec.rb
index 8aadfcb439b..dae546a0cdc 100644
--- a/spec/models/project_spec.rb
+++ b/spec/models/project_spec.rb
@@ -800,32 +800,14 @@ describe Project, models: true do
       end
 
       create(:note_on_commit, project: project2)
-    end
-
-    describe 'without an explicit start date' do
-      subject { described_class.trending.to_a }
 
-      it 'sorts Projects by the amount of notes in descending order' do
-        expect(subject).to eq([project1, project2])
-      end
+      TrendingProject.refresh!
     end
 
-    describe 'with an explicit start date' do
-      let(:date) { 2.months.ago }
+    subject { described_class.trending.to_a }
 
-      subject { described_class.trending(date).to_a }
-
-      before do
-        2.times do
-          # Little fix for special issue related to Fractional Seconds support for MySQL.
-          # See: https://github.com/rails/rails/pull/14359/files
-          create(:note_on_commit, project: project2, created_at: date + 1)
-        end
-      end
-
-      it 'sorts Projects by the amount of notes in descending order' do
-        expect(subject).to eq([project2, project1])
-      end
+    it 'sorts projects by the amount of notes in descending order' do
+      expect(subject).to eq([project1, project2])
     end
 
     it 'does not take system notes into account' do
diff --git a/spec/models/trending_project_spec.rb b/spec/models/trending_project_spec.rb
new file mode 100644
index 00000000000..cc28c6d4004
--- /dev/null
+++ b/spec/models/trending_project_spec.rb
@@ -0,0 +1,56 @@
+require 'spec_helper'
+
+describe TrendingProject do
+  let(:user) { create(:user) }
+  let(:public_project1) { create(:empty_project, :public) }
+  let(:public_project2) { create(:empty_project, :public) }
+  let(:public_project3) { create(:empty_project, :public) }
+  let(:private_project) { create(:empty_project, :private) }
+  let(:internal_project) { create(:empty_project, :internal) }
+
+  before do
+    3.times do
+      create(:note_on_commit, project: public_project1)
+    end
+
+    2.times do
+      create(:note_on_commit, project: public_project2)
+    end
+
+    create(:note_on_commit, project: public_project3, created_at: 5.weeks.ago)
+    create(:note_on_commit, project: private_project)
+    create(:note_on_commit, project: internal_project)
+  end
+
+  describe '.refresh!' do
+    before do
+      described_class.refresh!
+    end
+
+    it 'populates the trending projects table' do
+      expect(described_class.count).to eq(2)
+    end
+
+    it 'removes existing rows before populating the table' do
+      described_class.refresh!
+
+      expect(described_class.count).to eq(2)
+    end
+
+    it 'stores the project IDs for every trending project' do
+      rows = described_class.order(id: :asc).all
+
+      expect(rows[0].project_id).to eq(public_project1.id)
+      expect(rows[1].project_id).to eq(public_project2.id)
+    end
+
+    it 'does not store projects that fall out of the trending time range' do
+      expect(described_class.where(project_id: public_project3).any?).to eq(false)
+    end
+
+    it 'stores only public projects' do
+      expect(described_class.where(project_id: [public_project1.id, public_project2.id]).count).to eq(2)
+      expect(described_class.where(project_id: [private_project.id, internal_project.id]).count).to eq(0)
+    end
+  end
+end
diff --git a/spec/workers/trending_projects_worker_spec.rb b/spec/workers/trending_projects_worker_spec.rb
new file mode 100644
index 00000000000..c3c6fdcf2d5
--- /dev/null
+++ b/spec/workers/trending_projects_worker_spec.rb
@@ -0,0 +1,11 @@
+require 'spec_helper'
+
+describe TrendingProjectsWorker do
+  describe '#perform' do
+    it 'refreshes the trending projects' do
+      expect(TrendingProject).to receive(:refresh!)
+
+      described_class.new.perform
+    end
+  end
+end