From: "lrlebron@..." <lrlebron@...>
Date: 2008-07-02T06:37:26+09:00
Subject: Re: Using hpricot to get tables

On Jul 1, 4:03 pm, Dan Diebolt <dandieb...@yahoo.com> wrote:
> [Note:  parts of this message were removed to make it a legal post.]
>
> >I would like to access each table individually
>
> doc.search returns an array even if there is only one match. The consturct you are using iterates through this array:
>
> doc.search(strPath) do |div|
>
> end
>
> if you capture the search results into a variable named "divs" you can index it like and array (because it is one)
>
> divs=doc.search(strPath)
>
> If you want to immediately start iterating you can do this:
>
> doc.search(strPath).each_with_index do |div,idiv|
>   puts idiv if idiv==2
> end
>
> I work with hpricot a lot and I find it is more productive to not use all the fancy ruby idioms to shorten your code as you are dealing with pages that are very fragile to parse when someone changes the page content.
>
> See code below
> ==============
> require 'hpricot'
> require 'open-uri'
>
> strLink ="http://www.sportsline.com/mlb/gamecenter/boxscore/MLB_20080331_ARI@CIN"
> strPath ="//div[@class='SLTables1']/div"
>
> doc = Hpricot(open(strLink))
> divs=doc.search(strPath)
>
> puts "#{divs[0].inner_text.slice(0..70)}\n\n"
> puts "#{divs[1].inner_text.slice(0..70)}\n\n"
> puts "#{divs[2].inner_text.slice(0..70)}\n\n"
> puts "#{divs[3].inner_text.slice(0..70)}\n\n"

This works. Will be very useful for future projects.

I ended up using the xpath for each table which also worked.

Thanks,

Luis