From 9ec5b9bab1b41cd2869b583d28afcffbb429d702 Mon Sep 17 00:00:00 2001
From: Giuseppe Bilotta <giuseppe.bilotta@gmail.com>
Date: Tue, 18 Sep 2007 23:40:42 +0000
Subject: first_html_par: make Hpricot handling more robust

---
 lib/rbot/core/utils/utils.rb | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'lib/rbot')

diff --git a/lib/rbot/core/utils/utils.rb b/lib/rbot/core/utils/utils.rb
index 335d0506..0582cd4b 100644
--- a/lib/rbot/core/utils/utils.rb
+++ b/lib/rbot/core/utils/utils.rb
@@ -513,7 +513,8 @@ module ::Irc
         if pre_h.nil?
           pre_h = Hpricot::Elements[]
           found_h = false
-          doc.root.search("*") { |e|
+          doc.search("*") { |e|
+            next if e.bogusetag?
             case e.pathname
             when /^h\d/
               found_h = true
@@ -558,7 +559,8 @@ module ::Irc
         # we don't need
         if by_span.nil?
           by_span = Hpricot::Elements[]
-          doc.root.search("*") { |el|
+          doc.search("*") { |el|
+            next if el.bogusetag?
             by_span.push el if el.pathname =~ AFTER_PAR_PATH and (el[:class] =~ AFTER_PAR_CLASS or el[:id] =~ AFTER_PAR_CLASS)
           }
           debug "other \#1: found: #{by_span.pretty_inspect}"
-- 
cgit v1.2.3