Merge branch 'rs-optimize-award_user_list-spec' into 'master'

Optimize the `award_user_list` helper spec

According to
https://gitlab.com/gitlab-org/gitlab-ce/issues/23034#note_16586657, each
test for this helper generated 1,833 queries.

Now we only generate stubbed records, and only as many as we need for
each test.

This also corrects a slight logic bug in the helper itself. When the
number of awards was greater than the limit (9 by default), _and_ the
current user was one of them, we actually included 10 names, including
"You", plus the remaining count. Now we return the correct number
regardless.

See merge request !6722

app/helpers/issues_helper.rb

@@ -113,14 +113,13 @@ module IssuesHelper
     end
   end
 
-  def award_user_list(awards, current_user)
+  def award_user_list(awards, current_user, limit: 10)
     names = awards.map do |award|
       award.user == current_user ? 'You' : award.user.name
     end
 
-    # Take first 9 OR current user + first 9
     current_user_name = names.delete('You')
-    names = names.first(9).insert(0, current_user_name).compact
+    names = names.insert(0, current_user_name).compact.first(limit)
 
     names << "#{awards.size - names.size} more." if awards.size > names.size
 

spec/helpers/issues_helper_spec.rb

@@ -63,28 +63,38 @@ describe IssuesHelper do
   end
 
   describe '#award_user_list' do
-    let!(:awards) { build_list(:award_emoji, 15) }
+    it "returns a comma-separated list of the first X users" do
+      user = build_stubbed(:user, name: 'Joe')
+      awards = Array.new(3, build_stubbed(:award_emoji, user: user))
 
-    it "returns a comma seperated list of 1-9 users" do
+      expect(award_user_list(awards, nil, limit: 3))
-      expect(award_user_list(awards.first(9), nil)).to eq(awards.first(9).map { |a| a.user.name }.to_sentence)
+        .to eq('Joe, Joe, and Joe')
     end
 
     it "displays the current user's name as 'You'" do
-      expect(award_user_list(awards.first(1), awards[0].user)).to eq('You')
+      user = build_stubbed(:user, name: 'Joe')
+      award = build_stubbed(:award_emoji, user: user)
+
+      expect(award_user_list([award], user)).to eq('You')
+      expect(award_user_list([award], nil)).to eq 'Joe'
     end
 
-    it "truncates lists of larger than 9 users" do
+    it "truncates lists" do
-      expect(award_user_list(awards, nil)).to eq(awards.first(9).map { |a| a.user.name }.join(', ') + ", and 6 more.")
+      user = build_stubbed(:user, name: 'Jane')
+      awards = Array.new(5, build_stubbed(:award_emoji, user: user))
+
+      expect(award_user_list(awards, nil, limit: 3))
+        .to eq('Jane, Jane, Jane, and 2 more.')
     end
 
-    it "displays the current user in front of 0-9 other users" do
+    it "displays the current user in front of other users" do
-      expect(award_user_list(awards, awards[0].user)).
+      current_user = build_stubbed(:user)
-        to eq("You, " + awards[1..9].map { |a| a.user.name }.join(', ') + ", and 5 more.")
+      my_award = build_stubbed(:award_emoji, user: current_user)
-    end
+      award = build_stubbed(:award_emoji, user: build_stubbed(:user, name: 'Jane'))
+      awards = Array.new(5, award).push(my_award)
 
-    it "displays the current user in front regardless of position in the list" do
+      expect(award_user_list(awards, current_user, limit: 2)).
-      expect(award_user_list(awards, awards[12].user)).
+        to eq("You, Jane, and 4 more.")
-        to eq("You, " + awards[0..8].map { |a| a.user.name }.join(', ') + ", and 5 more.")
     end
   end