#!/usr/bin/ruby -I/home/tschroed/lib/ruby

# Licensed under the GNU General Public License 2.0 or greater
#
# Trevor Schroeder <tschroed@zweknu.org>
# Last Change: 8/2/2002
# 
# You'll need REXML to make this go.

require 'net/http'
require 'rexml/document'
require 'cgi'

require 'rsshtml'

###
### This would be the part you change
###

# Max word length before it's broken up. Set to 0 to not split
# words at any length.
MAXWORD=15

urls =	{
  'Boston' => 'http://www.moreover.com/cgi-local/page?o=rss&c=Boston%20news',
  'Nation' => 'http://trainedmonkey.com/news/rss.php?s=22',
  'Slashdot' => 'http://slashdot.org/slashdot.rss',
  'Blogdex' => 'http://blogdex.media.mit.edu/xml/fresh.asp?c=25',
  'Boing Boing' => 'http://boingboing.net/rss.xml',
  'Memepool' => 'http://memepool.com/memepool.rss',
  'Newsvac' => 'http://newsforge.com/newsvac.rss',
  'Newsforge' => 'http://newsforge.com/reports.rss'
  }

outfilename = '/home/tschroed/public_html/blog/import/sidebar.html'
rsshtml = []

## Grab Google News
# Googl dropped in favor of newsisfree feeds of the Washington Post and AP
# Newswire
#g="<h1 class=\"rssfeed\">google</h1>"
#conn = Net::HTTP.new('news.google.com',80)
#conn.start
#resp, rawdata = conn.get('/',nil)
#rawdata.each do |ln|
#	# Get the headline
#	ln=~/class=g href=(http:\/\/www.google.com\/url[^ ">]*)"?[^>]*>([^<]*)</
#	# Format!
#	g+="<p/><a href=\"#$1\" target=\"_blank\">#$2</a>\n" if($1!=nil)
#end
#rsshtml.push g

# Method for formatting individual items.
# Split up long words, strip out HTML tags (because it doesn't work in the
# tooltips window), put into a nice looking link.
def formatItem(element)
	# Get title, link, and description info for this item.
	t0=element.elements["title"][0].to_s
	lnk=element.elements["link"][0].to_s
	d0=element.elements["description"]

	# strip out HTML tags, etc from the description
	if(d0 != nil)
		d1=d0[0].to_s
		d1=CGI.unescapeHTML(d1)
		d1.gsub!('<[^>]*>','')
		dsc=CGI.escapeHTML(d1)
	else
		dsc=''
	end

	# Split the title if words are too long.
	t1=''
	t0.split(' ').each do |word|
		if MAXWORD != 0 && word.length > MAXWORD
			0.upto(word.length/MAXWORD) do |n|
				t1+= word.slice(n*MAXWORD,MAXWORD)+' '
			end
		else
			t1 += word+' '
		end
	end
	ttl=CGI.escapeHTML(t1)

	# and format it.
	return "<p/><a href=\"#{lnk}\" "+ ((dsc!='')?"title=\"#{dsc}\" ":"")+
	       "target=\"_blank\">#{ttl}</a>\n" 
end


# And now the rest
urls.sort.each do |title,url|

	rsshtml.push "<h1 class=\"rssfeed\">#{title}</h1>\n" +
	             # Note the use of the formatting block in the call to
		     # getHTML
	             RSSHTML.new(title, url).getHTML { |it| formatItem(it) }
end

outfile = File.open(outfilename,"w")
rsshtml.each { |text| outfile.puts text }
outfile.close