require 'spec_helper'

describe ReverseAdoc::Cleaner do
  let(:cleaner) { ReverseAdoc::Cleaner.new }

  describe '#scrub_whitespace' do
    it "makes consistent nonbreaking spaces" do
      result = cleaner.scrub_whitespace("     ")
      expect(result).to eq "     "
    end

    it "makes four linebreaks into two" do
      result = cleaner.scrub_whitespace("A\n\n\n\nB")
      expect(result).to eq "A\n\nB"
    end
  end

  describe '#clean_headings' do
    it "removes empty headings" do
      result = cleaner.clean_headings("<h2></h2>")
      expect(result).to eq " "
    end

    it "cleans superscripts rendered as headings" do
      result = cleaner.clean_headings(%{<p class="Standard" style="font-size: 12pt; font-family: Calibri; writing-mode: lr-tb; margin: 0;" align="left ! important"><span class="T1" style="margin: 0;">H</span><h1 class="T2" style="vertical-align: super; font-size: 58%; margin: 0;">2</h1><span class="T1" style="margin: 0;">0</span></p>})
      expect(result).to eq %{<p class="Standard" style="font-size: 12pt; font-family: Calibri; writing-mode: lr-tb; margin: 0;" align="left ! important"><span class="T1" style="margin: 0;">H</span><sup>2</sup><span class="T1" style="margin: 0;">0</span></p>}
    end
  end

  describe '#remove_newlines' do
    it 'removes more than 2 subsequent newlines' do
      result = cleaner.remove_newlines("foo\n\n\nbar")
      expect(result).to eq "foo\n\nbar"
    end

    it 'skips single and double newlines' do
      result = cleaner.remove_newlines("foo\nbar\n\nbaz")
      expect(result).to eq "foo\nbar\n\nbaz"
    end
  end

  describe '#remove_inner_whitespaces' do
    it 'removes duplicate whitespaces from the string' do
      result = cleaner.remove_inner_whitespaces('foo  bar')
      expect(result).to eq "foo bar"
    end

    it 'performs changes for multiple lines' do
      result = cleaner.remove_inner_whitespaces("foo  bar\nbar  foo")
      expect(result).to eq "foo bar\nbar foo"
    end

    it 'keeps leading whitespaces' do
      result = cleaner.remove_inner_whitespaces("    foo  bar\n    bar  foo")
      expect(result).to eq "    foo bar\n    bar foo"
    end

    it 'keeps trailing whitespaces' do
      result = cleaner.remove_inner_whitespaces("foo  \n")
      expect(result).to eq "foo  \n"
    end

    it 'keeps trailing newlines' do
      result = cleaner.remove_inner_whitespaces("foo\n")
      expect(result).to eq "foo\n"
    end

    it 'removes tabs as well' do
      result = cleaner.remove_inner_whitespaces("foo\t \tbar")
      expect(result).to eq "foo bar"
    end

    it 'keeps lines that only contain whitespace' do
      result = cleaner.remove_inner_whitespaces("foo \nbar \n \n  \nfoo")
      expect(result).to eq "foo \nbar \n \n  \nfoo"
    end
  end

  describe '#clean_punctuation_characters' do
    it 'removes whitespace between tag end and punctuation characters' do
      input = "**fat** . ~~strike~~ ? __italic__ ! "
      result = cleaner.clean_punctuation_characters(input)
      expect(result).to eq "**fat**. ~~strike~~? __italic__! "
    end
  end

  describe '#clean_tag_borders' do
    context 'with default_border is set to space' do
      before { ReverseAdoc.config.tag_border = ' ' }

      it 'removes not needed whitespaces from strong tags' do
        input = "foo ** foobar ** bar"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "foo **foobar** bar"
      end

      it 'remotes leading or trailing whitespaces independently' do
        input = "1 **fat ** 2 ** fat** 3"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 **fat** 2 **fat** 3"
      end

      it 'adds whitespaces if there are none' do
        input = "1**fat**2"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 **fat** 2"
      end

      it "doesn't add whitespaces to underscore'ed elements if they are part of links" do
        input = "![im__age](sou__rce)"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "![im__age](sou__rce)"
      end

      it "still cleans up whitespaces that aren't inside a link" do
        input = "now __italic __with following [under__scored](link)"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "now __italic__ with following [under__scored](link)"
      end

      it 'cleans italic stuff as well' do
        input = "1 __italic __ 2 __ italic__ 3__italic __4"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 __italic__ 2 __italic__ 3 __italic__ 4"
      end

      it 'cleans strikethrough stuff as well' do
        input = "1 ~~italic ~~ 2 ~~ italic~~ 3~~italic ~~4"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 ~~italic~~ 2 ~~italic~~ 3 ~~italic~~ 4"
      end
    end

    context 'with default_border set to no space' do
      before { ReverseAdoc.config.tag_border = '' }

      it 'removes not needed whitespaces from strong tags' do
        input = "foo ** foobar ** bar"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "foo **foobar** bar"
      end

      it 'remotes leading or trailing whitespaces independently' do
        input = "1 **fat ** 2 ** fat** 3"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 **fat** 2 **fat** 3"
      end

      it 'adds whitespaces if there are none' do
        input = "1**fat**2"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1**fat**2"
      end

      it "doesn't add whitespaces to underscore'ed elements if they are part of links" do
        input = "![im__age](sou__rce)"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "![im__age](sou__rce)"
      end

      it "still cleans up whitespaces that aren't inside a link" do
        input = "now __italic __with following [under__scored](link)"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "now __italic__with following [under__scored](link)"
      end

      it 'cleans italic stuff as well' do
        input = "1 __italic __ 2 __ italic__ 3__italic __4"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 __italic__ 2 __italic__ 3__italic__4"
      end

      it 'cleans strikethrough stuff as well' do
        input = "1 ~~italic ~~ 2 ~~ italic~~ 3~~italic ~~4"
        result = cleaner.clean_tag_borders(input)
        expect(result).to eq "1 ~~italic~~ 2 ~~italic~~ 3~~italic~~4"
      end
    end
  end

end