From: Jesse Crockett <tefflox@...>
Date: 2008-11-01T22:38:22+09:00
Subject: Please correct my Hpricot troubles.

Hello, I need some help using hpricot.  I'm trying simply to print out
the each link within "ul.class-name" for a CSV file.

ref: http://code.whytheluckystiff.net/hpricot/wiki/HpricotCssSearch

sample source for search "butter oil":

<ul class='mw-search-results'>

<li><a href="/wiki/Coconut_oil" title="Coconut oil">Coconut <span
class='searchmatch'>oil</span></a> ...

<li><a href="/wiki/Cocoa_butter" title="Cocoa butter">Cocoa <span
class='searchmatch'>butter</span></a></li>

</ul>


ruby:

search = "corn+flakes
bananas
milk+lowfat
blueberries+raw"

search = search.split("\n")

a = 0; until a == 3

  query = search[a]

  doc =
Hpricot(URI.parse("http://en.wikipedia.org/wiki/Special:Search?search=#{query}&fulltext=Search").read)

  # grab report list to build from (I need an array of inner html per
link)
  doc = (doc/"ul.mw-search-results a")
  line = "#{a + 1}|#{query}|"

  begin

    #
      line << ???  # fails for all attempts, please help here
    #

  rescue
    printf "%i|%s\n", a + 1, "## Exception Caught ##"
  end
  sleep 8 # + rand(20) # to avoid
  a += 1
end
-- 
Posted via http://www.ruby-forum.com/.