Commit 20eb0ab8 authored by Alex Kalderimis's avatar Alex Kalderimis

Place relative_positioning specs into the lib tree

This divides up the relative_positioning class specs into the
appropriate places in the lib tree
parent ccec75e5
# frozen_string_literal: true
require 'spec_helper'
RSpec.describe Gitlab::RelativePositioning::ItemContext do
let_it_be(:default_user) { create_default(:user) }
let_it_be(:project, reload: true) { create(:project) }
def create_issue(pos)
create(:issue, project: project, relative_position: pos)
end
# Increase the range size to convice yourself that this covers ALL arrangements
range = (101..108)
indices = (0..).take(range.size)
let(:start) { ((range.first + range.last) / 2.0).floor }
let(:subjects) { issues.map { |i| described_class.new(i.reset, range) } }
context 'there are gaps at the start and end' do
let_it_be(:issues) { (range.first.succ..range.last.pred).map { |pos| create_issue(pos) } }
it 'is always possible to find a gap' do
expect(subjects)
.to all(have_attributes(find_next_gap_before: be_present, find_next_gap_after: be_present))
end
where(:index) { indices.reverse.drop(2) }
with_them do
subject { subjects[index] }
it 'is possible to shift_right, which will consume the gap at the end' do
subject.shift_right
expect(subject.find_next_gap_after).not_to be_present
end
it 'is possible to shift_left, which will consume the gap at the start' do
subject.shift_left
expect(subject.find_next_gap_before).not_to be_present
end
end
end
context 'there is a gap of multiple spaces' do
let_it_be(:issues) { [range.first, range.last].map { |pos| create_issue(pos) } }
it 'is possible to find the gap from the right' do
gap = Gitlab::RelativePositioning::Gap.new(range.last, range.first)
expect(subjects.last).to have_attributes(
find_next_gap_before: eq(gap),
find_next_gap_after: be_nil
)
end
it 'is possible to find the gap from the left' do
gap = Gitlab::RelativePositioning::Gap.new(range.first, range.last)
expect(subjects.first).to have_attributes(
find_next_gap_before: be_nil,
find_next_gap_after: eq(gap)
)
end
end
context 'there are several free spaces' do
let_it_be(:issues) { range.select(&:even?).map { |pos| create_issue(pos) } }
let_it_be(:gaps) do
range.select(&:odd?).map do |pos|
rhs = pos.succ.clamp(range.first, range.last)
lhs = pos.pred.clamp(range.first, range.last)
{
before: Gitlab::RelativePositioning::Gap.new(rhs, lhs),
after: Gitlab::RelativePositioning::Gap.new(lhs, rhs)
}
end
end
where(:current_pos) { range.select(&:even?) }
with_them do
let(:subject) { subjects.find { |s| s.relative_position == current_pos } }
it 'finds the closest gap' do
closest_gap_before = gaps
.map { |gap| gap[:before] }
.select { |gap| gap.start_pos <= subject.relative_position }
.max_by { |gap| gap.start_pos }
closest_gap_after = gaps
.map { |gap| gap[:after] }
.select { |gap| gap.start_pos >= subject.relative_position }
.min_by { |gap| gap.start_pos }
expect(subject).to have_attributes(
find_next_gap_before: closest_gap_before,
find_next_gap_after: closest_gap_after
)
end
end
end
context 'there is at least one free space' do
where(:free_space) { range.to_a }
with_them do
let(:issues) { range.reject { |x| x == free_space }.map { |p| create_issue(p) } }
let(:gap_rhs) { free_space.succ.clamp(range.first, range.last) }
let(:gap_lhs) { free_space.pred.clamp(range.first, range.last) }
it 'can always find a gap before if there is space to the left' do
expected_gap = Gitlab::RelativePositioning::Gap.new(gap_rhs, gap_lhs)
to_the_right_of_gap = subjects.select { |s| free_space < s.relative_position }
expect(to_the_right_of_gap)
.to all(have_attributes(find_next_gap_before: eq(expected_gap)))
end
it 'can always find a gap after if there is space to the right' do
expected_gap = Gitlab::RelativePositioning::Gap.new(gap_lhs, gap_rhs)
to_the_left_of_gap = subjects.select { |s| s.relative_position < free_space }
expect(to_the_left_of_gap)
.to all(have_attributes(find_next_gap_after: eq(expected_gap)))
end
end
end
end
This diff is collapsed.
This diff is collapsed.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment