Does ruby strip headers from response? - ruby-on-rails

Question

I am fetching html content directly from my blog as:
response = Net::HTTP.get_response(uri)
respond_to do |format|
format.html { render :text => response.body }
end
Although at the blog engine (WordPress) I am adding header Access-Control-Allow-Origin: * how ever I noticed that its not passed within the response.
However, if I use postman to get the page or view the page into browser directly, I can see that the header is there.
EDIT
I can see other headers passed, ex:
cache-control: no-cache, must-revalidate, max-age=0
content-type: text/html; charset=UTF-8
date: Tue, 24 Jul 2018 06:37:57 GMT
expires: Wed, 11 Jan 1984 05:00:00 GMT
Any idea?

Answer 1

response.body will return you body part not header part. you can convert response to hash and check header like below:
> url = "https://stackoverflow.com/questions/51492025/does-ruby-strip-headers-from-response"
> uri = URI.parse(url)
> response = Net::HTTP.get_response(uri)
#=> #<Net::HTTPOK 200 OK readbody=true>
> response.to_hash
#=> {"cache-control"=>["private"], "content-type"=>["text/html; charset=utf-8"], "last-modified"=>["Tue, 24 Jul 2018 07:04:00 GMT"], "x-frame-options"=>["SAMEORIGIN"], "x-request-guid"=>["22a4b6b6-3039-46e2-b4de-c8af7cad6659"], "strict-transport-security"=>["max-age=15552000"], "content-security-policy"=>["upgrade-insecure-requests"], "accept-ranges"=>["bytes", "bytes"], "age"=>["0", "0"], "content-length"=>["31575"], "date"=>["Tue, 24 Jul 2018 07:04:46 GMT"], "via"=>["1.1 varnish"], "connection"=>["keep-alive"], "x-served-by"=>["cache-bom18221-BOM"], "x-cache"=>["MISS"], "x-cache-hits"=>["0"], "x-timer"=>["S1532415886.990199,VS0,VE280"], "vary"=>["Accept-Encoding,Fastly-SSL"], "x-dns-prefetch-control"=>["off"], "set-cookie"=>["prov=a7dfe911-76a1-f1c1-093b-3fc8fe79af65; domain=.stackoverflow.com; expires=Fri, 01-Jan-2055 00:00:00 GMT; path=/; HttpOnly"]}
You can access specific header as below by passing header name:
> response['Cache-Control']
#=> "private"
for more details read: https://ruby-doc.org/stdlib-2.5.1/libdoc/net/http/rdoc/Net/HTTP.html

Answer 2

If you want to pass through headers that are being served from the host that you're fetching from, you first need to stash the response from your blog in a different variable name. Let's call it blog_response (this is because response is a preexisting special method name in a rails controller instance.).
blog_response = Net::HTTP.get_response(uri)
Then you need to grab the header you care about from the blog_response like this:
header_name, header_value = blog_response.each_header.find do |name, value|
name =~ /pattern-matching-a-header-name-i-care-about/i #case insensitive regex matching recommended for http headers
end
Then you need to set them in your controller before you render the response, e.g.:
response.headers[header_name] = header_value
respond_to do |format|
format.html { render :text => blog_response.body }
end
This example is obviously only for one header, but you can copy multiple headers by just iterating through, matching and setting them in your response like so:
blog_response.each_header.select do |name, value|
if name =~ /pattern-matching-header-names-i-care-about|some-other-pattern-i-care-about/i #case insensitive regex matching recommended for http headers
response.headers[name] = value
end
end
If you want to pass all headers through just do this:
blog_response.each_header do |name, value|
response.headers[name] = value
end
respond_to do |format|
format.html { render :text => blog_response.body }
end

Answer 3

Net::HTTPResponse (that is your response) mixes in Net::HTTPHeader. Thus, you can get an individual header as response['Access-Control-Allow-Origin'], iterate over them with response.each_header, or even get them all as a hash using response.to_hash.