From: Peter Fitzgibbons <peter.fitzgibbons@...>
Date: 2005-10-26T00:44:56+09:00
Subject: Re: REXML::XPath results out of sort order ?

------=_Part_8666_31922828.1130255093094
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

On 10/21/05, daz <dooby@d10.karoo.co.uk> wrote:
>
>
> Peter Fitzgibbons wrote:
> > Peter Fitzgibbons wrote:
> > >
> > > How do I find my current REXML version ? I'm using the default lib
> > >installed with one-click 1.8.15
> >
> > Oh, yes, I'm using 3.1.3.
>
>
> There's a contradiction, somewhere (?)
>
> "One-click" 1.8.2-15 Stable Release (April 20, 2005) has REXML 3.1.2.1<ht=
tp://3.1.2.1>
> in Ruby 1.8.2 final (2004-12-25).
>
>
> require 'rexml/document'
> puts 'ruby %s (%s) [%s]' % [RUBY_VERSION, RUBY_RELEASE_DATE,
> RUBY_PLATFORM]
> puts REXML::Version
>
> xml =3D REXML::Document.new(File.read
> ('D:\ruby\DfB\TEMP\clruby\___test.xml'))
> mi =3D REXML:: XPath.match(xml,
> '//HomesiteTree/Builder/Division/Workflow/Milestone')
>
> mi.each { |e| print ' ', e.elements['MsMilestoneId'].text }; puts
>
> #------------------------------------------
>
> #-> ruby 1.8.2 (2004-12-25) [i386-mswin32]
> #-> 3.1.2.1 <http://3.1.2.1>
> #-> 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210
> 220 230 240 270 280 290 10 20
>
> #------------------------------------------
>
> #-> ruby 1.8.3 (2005-07-22) [i386-mswin32]
> #-> 3.1.3
> #-> 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 20=
0
> 210 220 230 240 270 280 290
>
> #------------------------------------------
>
>
> > So what's different then ?
>
> I *know* I'm using 3.1.3 :)
>
> It fixes several bugs relating to XPath result ordering.
> The REXML tracker is down as I write, otherwise I'd give a link.
>
> If you're sure you're getting incorrect results using 3.1.3,
> please submit a bug report to:
> http://www.germane-software.com/projects/rexml/
>
> As you can see from my tests - the 2nd result array is in order.
>
> The disorder of the 1st result seems to be different from yours
> (first element of <80>) but maybe there's a random factor
> associated with the bug in 3.1.2.1 <http://3.1.2.1> (?)
>
>
> daz
>
>
>
>
> #-> ruby 1.8.2 (2004-12-25) [i386-mswin32]
#-> 3.1.2.1 <http://3.1.2.1>
Yes, this is my result.
I may have to update my system with ruby 1.8.3. On windows. Argh.

------=_Part_8666_31922828.1130255093094--