From: gf <8si.greg@...>
Date: 2010-04-06T14:45:28+09:00
Subject: Re: how to use an array to load html data with ruby

Here's a similar way to do what Josh does, only using Nokogiri:


#!/usr/bin/ruby

require 'rubygems'
require 'nokogiri'

HTML =<<EOT
<table border=1>
  <tr>
    <td>report_data</td>
    <td><strong>liu_asset</strong></td>
    <td><strong> cash</strong></td>
    <td><strong>finance_asset</strong></td>
    <td><strong> note</strong></td>
  </tr>
  <tr>
    <td>2009-12-31</td>
    <td>0 </td>
    <td>1,693,048,000,000</td>
    <td>20,147,000,000</td>
    <td>500</td>
  </tr>
  <tr>
    <td>2009-09-30</td>
    <td>0 </td>
    <td>1,777,512,000,000</td>
    <td>24,977,000,000</td>
    <td>700</td>
  </tr>
</table>
EOT

doc = Nokogiri::HTML.parse(HTML)

array = []
doc.css('tr').each_with_index do |tr, tr_i|
  array[tr_i] = tr.css('td').map{ |td| td.text }
end

array[0][0] # => "report_data"
array[0][1] # => "liu_asset"
array[0][2] # => " cash"
array[0][3] # => "finance_asset"
array[0][4] # => " note"
array[1][0] # => "2009-12-31"
array[1][1] # => "0 "
array[1][2] # => "1,693,048,000,000"
array[1][3] # => "20,147,000,000"
array[1][4] # => "500"
array[2][0] # => "2009-09-30"
array[2][1] # => "0 "
array[2][2] # => "1,777,512,000,000"
array[2][3] # => "24,977,000,000"
array[2][4] # => "700"