From: Aaron Patterson <aaron@...>
Date: 2008-09-18T13:27:15+09:00
Subject: Re: How can one get the Hpricot DOM document from Mechanize?

On Sun, Sep 14, 2008 at 04:03:04AM +0900, Just Another Victim of the Ambient Morality wrote:
>     I was wondering if there were some way of getting the Hpricot DOM (for 
> lack of a better term) from a Mechanize page.  For example:
> 
> 
> agent = WWW:Mechanize.new
> page = agent.get(http://www.website.com)
> 
> # I am currently doing this
> doc = Hpricot(page.body)
> 
> # I would like to do this
> doc = page.get_hpricot_dom
> 
> 
>     The idea is that since Mechanize apparently uses Hpricot and it's surely 
> using it to parse the HTML begotten from the agent.get method, it would be 
> nice if I didn't have to repeat that work.
>     Is there a way to get this Hpricot document? ...or am I just totally 
> wrong about how Mechanize uses Hpricot?

You can get at the Hpricot document by using the "parser" accessor on
WWW::Mechanize::Page.  Page also responds to "search", "/", and "at",
which just delegate to the Hpricot document.

So you can just do:

(agent.get('http://tenderlovemaking.com')/'tr').each do |tr|
  ...
end

-- 
Aaron Patterson
http://tenderlovemaking.com/