Rspec testing for html entities in page content

Question

I'm writing a request spec and would like to test for the presence of the string "Reports » Aging Reports". I get an error (invalid multibyte char) if I put in the character in my matcher expression directly, so I tried this: page.should have_content("Reports » Aging Reports")

This fails the test with the message:

expected there to be content "Reports » Aging Reports" in "\n Reports » Aging Reports\n

I've tried things like .html_safe with no success. Is there a way to test for text containing html entities?

Edit:

Here's the relevant area of the html source:

<a href="/reports">Reports</a> &raquo; <a href="/aging_reports">Aging Reports</a>

Answer 1

You can have the test look at the raw source of your page too:

breadcrumb = '<a href="/reports">Reports</a> &raquo; <a href="/aging_reports">Aging Reports</a>'
page.body.should include(breadcrumb)
expect(page.body).to include(breadcrumb) # rspec 2.11+

With that said I'm not sure that's the most elegant solution. Assuming there is a class named .breadcrumb around your links you could just validate the links are present within the div:

within '.breadcrumb' do
  page.should have_css('a', text: 'Reports')
  page.should have_css('a', text: 'Aging Reports')
  expect(page).to have_css('a', text: 'Reports') # rspec 2.11+
  expect(page).to have_css('a', text: 'Aging Reports') # rspec 2.11+
end

This way you are explicitly looking for a href's within the breadcrumb block.

Answer 2

html_safe does not remove html tags if thats what you want.

If you will remove html tags before you test you can use sanitize gem.

# in Gemfile
gem 'sanitize'

# in testfile
require 'sanitize'
html = Sanitize.clean(page.body.text)
html.should have_content("Reports » Aging Reports")

Answer 3

Sometimes the simplest solution is the right one and things just work...

page.should have_content("Reports » Aging Reports")

Answer 4

I found a workaround for the time being that uses a regex instead of a string:

find('#breadcrumbs').text.should match(/Reports . Aging Reports/)

This gets the text of the breadcrumbs div, which contains the text I'm looking for, so the scope of the match is limited. This works fine, but I'd still love to know how to match specific html entities.