This should work on all the Index, Dow Jones, NASDAQ Qs….

So here’s the paste of the code. The idea is very similar to Dog of FTSE, except this uses the intelligent learning feature of scRubyt instead of the straight XPATH. This feature of scRubyt is especially useful when it is used on a well-element-classified (don’t know how to phrase it) site like yahoo finance.

#
# Dog of HSI written by Houston Ng

require ‘rubygems’
require ’scrubyt’

# Initialize an empty hash for final data
final_data={}

# Extractor 1: Get an update symbol list of HSI from yahoo finance
hsi_list = Scrubyt::Extractor.define do
fetch ‘http://hk.finance.yahoo.com/q/cp?s=%5EHSI’
top “//td[@class=’c1′]” do
symbol “//a”
end
end
hsi_stocks = hsi_list.to_hash

# Extractor 2: Get the stock information, just modify the pattern learner
hsi_stocks.each do |hsi_stock|
sc = hsi_stock[:symbol]
stock_data = Scrubyt::Extractor.define do
fetch ‘http://hk.finance.yahoo.com/q’
fill_textfield ’s’, sc
submit

#For the div with id, stock bar,
#get the content inside h3, the first and second b tags
#This only works for HK finance
top “//div[@id=’stock-bar’]” do
name “//h3″
price “//b[1]”
change “//b[2]”
end
end
#Save the stock data into the final data
final_data[sc] = stock_data.to_hash
end
# Print out the final data, basically just puts out the hash on screen
#The example is just spitting out the name for simplicity
final_data.each do |key, entry|
puts “\n#{key}”
entry.each do |datapair|
puts “#{datapair[:name]} | Price: #{datapair[:price]} | Change: #{datapair[:change]}”
end
end
I know I can just get everything from the same page I got the HSI symbol list. But Just for fun to play with auto-fill in. I load the actual quote page.

There is however, a slight pitfall that my little brain needs to work around. If you notice the output, there are two empty hashes. They are from the HSI list grabbing. It grabs too many cells, that it grab the header cell but since, the header cell does not contain any <a> tags, it return an empty value. This will be next improvement of the code…

Here’s the source code Dog of HSI

Dog of the Dow coming soon….

Well, Thanks to Apple switching to Intel. We have parallels, which leads to IE6 or IE7… depending which one which one you’ve installed and there can only be one installed at the same time humm…

Here’s the solution.

I just found out how you can install all the IE from 3.3 to 6 all in a nice package HA!

Multiple IE from Tredo Soft

1. SVN. Used to use it back at my old aero company. It’s great but the code pretty much sits in the companies’ server. That was the old day.. Anyway, There are tons of these online hosting too! There are, Unfuddle and Assembla
2. GIT. New “IT” thing for the Web2.0 community. And now it’s much cleaner and simpler with the new GitHub repository service. Now code can stay online and shared anywhere! Great! Pricing is not too bad for a startup too like $7-12/month. Umm… sound like I need to get an account there… wait I don’t have any code nor collaborators…. well versioning is still a good habit