Question

Not sure why I’m getting the following error when the URI works just fine in the browser:

http://oracleofbacon.org/cgi-bin/xml?a=Kevin Bacon&b=Tom Cruise&u=1&p=google-apps

This is my code:

def kb(to)

    uri = "http://oracleofbacon.org/cgi-bin/xml?a=Kevin Bacon&b=#{to.strip}&u=1&p=google-apps"
    doc = Nokogiri::XML(open(uri)) # throws error on this line

    return parse(doc)
end

I get the following error:

in `split': bad URI(is not URI?): http://oracleofbacon.org/cgi-bin/xml?a=Kevin Bacon&b=Tom Cruise&u=1&p=google-apps (URI::InvalidURIError)`

I execute the method in the following way:

kb("Tom Cruise")

Answer 1

It’s because a browser is pathologically friendly, like a puppy, and will go to great lengths to render a page or resolve a URL. An application won’t do that because you have to tell it how to be friendly.

Your URL is not valid because it has embedded spaces. Replace the spaces with %20:

irb -f
irb(main):001:0> require 'open-uri'
=> true
irb(main):002:0> open('http://oracleofbacon.org/cgi-bin/xml?a=Kevin%20Bacon&b=Tom%20Cruise&u=1&p=google-apps').read
=> "<?xml version=\"1.0\" standalone=\"no\"?>\n<link><actor>Tom Cruise</actor><movie>A Few Good Men (1992)</movie><actor>Kevin Bacon</actor></link>"

---

Escaping the characters needing to be escaped is easy:

irb -f
irb(main):001:0> require 'uri'
=> true
irb(main):002:0> URI.escape('http://oracleofbacon.org/cgi-bin/xml?a=Kevin Bacon&b=Tom Cruise&u=1&p=google-apps')
=> "http://oracleofbacon.org/cgi-bin/xml?a=Kevin%20Bacon&b=Tom%20Cruise&u=1&p=google-apps"