From: Jan Svitok <jan.svitok@...>
Date: 2007-03-05T21:14:19+09:00
Subject: Re: ruby noob

On 3/5/07, mosfet <richom.v@free.fr> wrote:
> Hi,
>
> I would like to parse a very simple html(index_msg.htm) file described
> below :
>
> <tr>
>    <td>WM_ACTIVATE</td>
>    <td>0x0006</td>
>    <td></td>
>    <td>0x0000</td>
>    <td>WM_NULL</td>
> </tr>
> <tr>
>    <td>WM_ACTIVATEAPP</td>
>    <td>0x001C</td>
>    <td></td>
>    <td>0x0001</td>
>    <td>WM_CREATE</td>
> </tr>
> ...
>   I would like to parse this file and to extract information like this :
>
> enum foo
> {
>    eWM_ACTIVATE = 0x0006,
>    eWM_ACTIVATEAPP = 0x0001,
>   ...
> };
>
> I am starting with this :
>
>
> fileIn  = File.open("C:/WIKI_CE/index_msg.htm", "r")
> fileOut = File.new("C:/WIKI_CE/enumWmMsg.h", "w")
>
> begin
>      while (line = fileIn.readline)
>          line.chomp
>          $stdout.print line
>      end
> rescue EOFError
>      fileIn.close
>      fileOut.close
> end
>
> but now I am stuck. Should I use regex or how can I compare two string ?

1. have a look at hpricot
2. if it's too big for you use regexen with /m flag, and use Regex#scan():

REGEX = /<tr>\s*
<td>(.*?)<\/td>\s*
<td>(.*?)<\/td>
<td>(.*?)<\/td>
<td>(.*?)</td>
<td>(.*?)</td>
<\/tr>/xm

file_in  = File.read("C:/WIKI_CE/index_msg.htm")
File.open("C:/WIKI_CE/enumWmMsg.h", "w") do |file_out|
      file_in.scan(REGEX) do
          file_out.puts $1, $2, $3, $4, $5
      end
    end
end

Notes:
1. we_use_snake_case_for_variable_names
2. Use File.open with block to automatically close the file
3. You'll have the values in $1..$5
4. It seems you are inconsistent - in the first example you chose the
second line, in the other the fourth one.

In any case, Peter's approach will be easier, and more stable.