How to use Regex expression in Capybara have_css/have_selector?

Question

In Rspec Capybara test I want to check that no H4, H5, ..., H9 selector exists.

I would expect the following solution:

visit "/"
expect(page).not_to have_css(/(h|H)\[4-9]d/)

But it fails due to Nokogiri parsing

I also noticed that:

has_css?("h1")

# is not equal to

has_css?("H1")

why not just "h1,h2,h3,h4,h5,h6,h7,h8,h9". I'm not aware that nokogiri takes regex selectors at all. If you are too lazy to type them out you could do `Array(1..9).map { |n| 'h' + n.to_s }.join(',')` — max, Apr 21 '15 at 10:03
@papirtiger This is what I do, but there should be a better way to test it. I think that has_css? should be case insensitive. — dpaluy, Apr 21 '15 at 10:07
How is that really an issue unless you are testing HTML from 1995? — max, Apr 21 '15 at 10:11
I should verify that specific page doesn't have H* tags (SEO requirement) — dpaluy, Apr 21 '15 at 11:02

score 1 · Accepted Answer · edited May 23 '17 at 12:22

Some alternatives:

1. Just create a slightly ugly expectation and get it done:

visit "/"
expect(page).not_to have_css("h1,H1,h2,H2,h3,H3,h4,H4,h5,H5,h6,H6,h7,H7,h8,H8,h9,H9")

I used the following snippet to generate that ugly selector:

(1..9).map { |n| "h#{n.to_s},H#{n.to_s}" }.join(',')

Nokogumbo is a HTML5 parser which creates lowercase element names.