# 2009 12 31 john blue
# ruby code to extract question elements for twitter discussions
# The questions/responses must have a string that has a 'q' and a number
require 'rubygems'
require 'hpricot'
require 'open-uri'
# Pull command line items: intro text, question number, web page url containing twitter discussion
# For ref on command line info see http://ruby.about.com/od/rubyfeatures/a/argv.htm
introtext = ARGV[0] # example "Agchaxt for 2009-12-29"
questionnumber = ARGV[1] # example 1
weburl = ARGV[2] # example "http://localhost:8888/files/Agchat20081229Frame.html"
# Get the Twitter discussion input web page info
doc = Hpricot(open(weburl))
#-----------
# get the head part of page, set absolute link, and update all relative links to absolute links
thehead=doc.search("/html/head")
thebody=doc.search("/html/body")
theresultsul=doc.search("/html/body/div/div/div/ul")
#-----------
# set up output
# beginning html file codes
htmlstart = ''
htmlhead='
'
# setup the file pre-html
puts htmlhead
########### Set the ul pre html
ulprehtml='
' + introtext +' Note: The posts are read from top to bottom.
'
# setup the file post-html
ulposthtml='
'
htmlclose= ''
# setup regular expression for use within ruby if statement
regexpression = "Q.? ?"+ "%d" % questionnumber
regexforif = Regexp.new(regexpression, true) # for details see Regexp.new at http://ruby-doc.org/core/classes/Regexp.html
puts htmlstart
puts ulprehtml
puts '
'
########### Set the questions to search
# see http://icfun.blogspot.com/2008/04/ruby-regular-expression-handling.html
# and use to test reg exp http://rubular.com/
liitems = theresultsul.search('//li')
liitems.map do |liitem|
limsg = liitem.search('//div[2]')
if (limsg.inner_html =~ regexforif )
puts liitem
end
end
puts '