I am trying to get assert_select to work on some raw text. Below is what I have, but when I perform the select it says there are no matches. I have also attached a snippet from the assert_select method which shows that accepting an HTML::Node object is doable (I have also tried wrapping the extracted html in html/body tags)
# Controller test
test "edit form contains all required fields" do
group = groups(:a_group)
user = users(:group_member)
post = create_post(user, :group_ids => group.id)
xhr :get, :edit, {:id => post.id}, {:user_id => user.id}
node = convert_js_to_node(@response.body, "form")
puts node.inspect
assert_select node, "form"
assert_select node, "input[type=text]"
end
# test_helper.rb method
def convert_js_to_node(js, enclosing_tag)
regex = Regexp.new("(\<#{enclosing_tag}.*#{enclosing_tag}\>)")
HTML::Node.new( js.match(regex)[0] )
end
## Test output
Loaded suite functional/posts_controller_test
Started
#<HTML::Node:0x1050e6040 @line=0, @children=[], @parent="<form accept-charset=\\\"UTF-8\\\" action=\\\"/posts/1023110311\\\" class=\\\"edit_post\\\" data-post-id=\\\"1023110311\\\" data-remote=\\\"true\\\" id=\\\"edit_post_1023110311\\\" method=\\\"post\\\"><div style=\\\"margin:0;padding:0;display:inline\\\"><input name=\\\"utf8\\\" type=\\\"hidden\\\" value=\\\"✓\\\" /><input name=\\\"_method\\\" type=\\\"hidden\\\" value=\\\"put\\\" /><\\/div>\\n\t\t<h3>Edit post<\\/h3>\\n\t\t<h5><img alt=\\\"Contact\\\" class=\\\"icon\\\" height=\\\"16\\\" src=\\\"/images/contact.png?1288409655\\\" width=\\\"16\\\" /> in some where<\\/h5>\\n\t\t\t\\n\t\t<div id=\\\"post_content\\\" class=\\\"clearfix clear\\\"
1) Failure:
test_edit_form_contains_all_required_fields(PostsControllerTest)
[]:
Expected at least 1 element matching "form", found 0.
<false> is not true.
# File actionpack/lib/action_controller/assertions/selector_assertions.rb, line 201
def assert_select(*args, &block)
# Start with optional element followed by mandatory selector.
arg = args.shift
if arg.is_a?(HTML::Node)
# First argument is a node (tag or text, but also HTML root),
# so we know what we're selecting from.
root = arg
arg = args.shift
elsif arg == nil
...
This is what I put together. It isn't perfect but it does what I need it to.
# allows for a method similar the native assert_select method to be performed
# on raw html text that is embedded within javascript code.
# ex.
# $('body').append("<div id=\"edit_post_form_container\" class=\"active modal_form_container\">
# <a href=\"#\" class=\"close\">
# <img alt=\"Close\" height=\"16\" src=\"/images/close.png?1293730609\" width=\"16\" />
# <\/a>
# <form accept-charset=\"UTF-8\" action=\"/posts/1023110335\" .../>")
#
# To test for the closing link image:
# => assert_js_select("img[alt=Close]")
# To be more specific you can include a parent tag:
# => assert_js_select("img[alt=Close]", "a")
#
# Note: The parent_tag parameter cannot include any attribute or style selectors since
# the initial extraction of the html section is done using a regular expression.
def assert_js_select(tag, content=nil, options={})
text = options[:text] || @response.body
parent_tag = options[:parent_tag] || ""
regex = Regexp.new("(\<#{parent_tag}.*#{parent_tag}\>)")
extracted_text = text.match(regex)[0].gsub(/\\/, "")
if extracted_text.present?
doc = HTML::Document.new(CGI::unescapeHTML(extracted_text)).root
assert_select(doc, tag, content)
else
assert false, "Missing #{tag} tag"
end
end