From: Paul Lutus <nospam@...>
Date: 2006-11-28T04:00:13+09:00
Subject: Re: How to get data from html table

Vikash Kumar wrote:

> I want to store the values of a table in different variables, I have the
> following table structure:
> 
> <table width="579">
>   <tr class="even">
>     <td class width="65">&nbsp;Case5-04</td>
>     <td class width="130">10/11/2006 23:24:33</td>
>     <td class width="61">Case5-04</td>
>     <td class width="32">1005</td>
>     <td class width="59">Sell</td>
>     <td class width="36">1,000</td>
>     <td class width="34">ARP</td>
>     <td class width="52">$36.90</td>
>   </tr>
>   <tr class="odd">
>     <td class width="65">&nbsp;Case5-03</td>
>     <td class width="130">10/11/2006 23:20:07</td>
>     <td class width="61">Case5-03</a></td>
>     <td class width="32">1005</td>
>     <td class width="59">Buy</td>
>     <td class width="36">1,500</td>
>     <td class width="34">ARP</td>
>     <td class width="52">$36.70</td>
>   </tr>
>   <tr class="even">
>     <td class width="65">&nbsp;Case4-04</td>
>     <td class width="130">10/11/2006 05:28:54</td>
>     <td class width="61">Case4-04</a></td>
>     <td class width="32">1004</td>
>     <td class width="59">Sell</td>
>     <td class width="36">300</td>
>     <td class width="34">RIL</td>
>     <td class width="52">$490.00</td>
>   </tr>
>   <tr class="odd">
>     <td class width="65">&nbsp;Case4-03</td>
>     <td class width="130">10/11/2006 05:21:32</td>
>     <td class width="61">Case4-03</a></td>
>     <td class width="32">1004</td>
>     <td class width="59">Buy</td>
>     <td class width="36">200</td>
>     <td class width="34">RIL</td>
>     <td class width="52">$489.90</td>
>   </tr>
> </table>
> 
> I want to store the values in variables so that I can compare records.
> Please help me out how to do this in ruby.

Only a few lines of Ruby are required to accomplish this: 

------------------------------------------

#!/usr/bin/ruby -w

data = File.read(sourcefilename)

output = []

html_rows = data.scan(%r{<tr.*?>(.*?)</tr>}im).flatten

html_rows.each do |row|
   # filter these undesired elements
   row.gsub!("&nbsp;","")
   row.gsub("</a>","")
   cells = row.scan(%r{<td.*?>(.*?)</td>}im).flatten
   output << cells
end

# done collecting, now display

output.each do |row|
   line = row.join(",")
   puts line
end

------------------------------------------

Output:

Case5-04,10/11/2006 23:24:33,Case5-04,1005,Sell,1,000,ARP,$36.90
Case5-03,10/11/2006 23:20:07,Case5-03</a>,1005,Buy,1,500,ARP,$36.70
Case4-04,10/11/2006 05:28:54,Case4-04</a>,1004,Sell,300,RIL,$490.00
Case4-03,10/11/2006 05:21:32,Case4-03</a>,1004,Buy,200,RIL,$489.90

BTW there are some errors in your HTML sample. One example are some orphan
"</a> half-tags. They are not difficult to filter out of the result.

One reason this sort of example is difficult to parse is because of the
errors. But this specific parser will do the basic parsing, and you can
always filter a few errors, as this code does.

Digression: when solving a problem like this, it is often much easier to
write a few lines of HTML than to try to use a high-powered library to
accomplish it.

-- 
Paul Lutus
http://www.arachnoid.com