From: Francis Hwang <sera@...>
Date: 2005-05-15T13:35:57+09:00
Subject: [ANN] HtmlClipping 0.1.0

Hi all,

I've just released the first version of HtmlClipping.

== What's HtmlClipping?

HtmlClipping is a library for reading an HTML page that has a link 
pointing to a particular URI. It removes most HTML markup, bolds the 
link text, and trims the resulting text to a fixed number of 
characters. I developed it to help me track referers to my website, 
though I suppose it might have other uses.

For example, the following script gets the HTML at 
rubyforge.org/credits/, and forms an excerpt around the link to 
www.rubycentral.org/pledge/.

   require 'htmlclipping'
   require 'net/http'

   contents = ''
   Net::HTTP.start( 'rubyforge.org' ) do |http|
     response = http.get '/credits/'
     contents = response.body
   end
   clipping = HtmlClipping.new(
     contents, 'http://www.rubycentral.org/pledge/', 500
   )
   puts clipping.to_s

   => "&#8230; RubyForge takes time, effort, and money. Many thanks to 
the
      folks listed below who are making it possible! <br /> If RubyForge 
has
      been helpful to you, and you want to give something back to the 
Ruby
      community, please consider supporting <strong>RubyCentral</strong>.
      Thanks! <br /> InfoEther, Inc purchased the RubyForge hardware and
      provides system administration support. <br /> Several folks 
provide
      file mirrors to help share the bandwidth load: <br /> Evan Webb 
<br />
      Dennis Oelkers <br /> Austin &#8230;"

http://rubyforge.org/projects/htmlclipping

Francis Hwang
http://fhwang.net/