From: Mark Thomas <mark@...>
Date: 2009-11-10T11:35:05+09:00
Subject: Re: Using Nokogiri

On Nov 9, 12:37 am, 7stud -- <bbxx789_0...@yahoo.com> wrote:
> Mark Thomas wrote:
> > This should get what you want:
>
> > prefix = '//div[@class="sectionHeaderText"]/following-sibling::'
> > xpaths = {
> >  :name => "#{prefix}b/text()",
> >  :addr => "#{prefix}text()[2]",
> >  :citystzip => "#{prefix}text()[3]",
> >  :country => "#{prefix}text()[4]",
> >  :phone => "#{prefix}text()[5]",
> > }  
> > xpaths.each do |data,xpath|
> >   puts "#{data} = " + doc.search(xpath).to_s.strip
> > end
>
> > -- Mark.
>
> I was wondering if you could answer some xpath questions? I would think
> that in this xpath:
>
> div[@class="sectionHeaderText"]/following-sibling::text()[2]
>
> the part:
>
> div[@class="sectionHeaderText"]/following-sibling
>
> would be the <b> tag.

Not quite. following-sibling:: is an axis predicate that needs to be
followed by a node. Therefore following-sibling::text() is a set of
all text nodes after the div. After that, it's just a matter of
indexing.

> doesn't seem applicable.  And in fact, when I run your code, it doesn't
> work:

As I just posted in another message, it works for me. I wonder what's
different about my environment. Are you using Nokogiri 1.4.0?