Re: [xml] xmlHashAddEntry/xmlHashScan behaviour

[Daniel Veillard <veillard redhat com>]

On Wed, Oct 31, 2012 at 09:22:17AM +0300, Nikolay Sivov wrote:
> On 10/31/2012 08:09, Daniel Veillard wrote:
> > On Wed, Oct 31, 2012 at 08:45:35AM +0300, Nikolay Sivov wrote:
> > > Hi,
> > >
> > > Got a question on libxml2 hash implementation.
> > > Let's say I want to have index access to hash data, so I call
> > >
> > > xmlHashAddEntry for item1
> > > xmlHashAddEntry for item2
> > > xmlHashAddEntry for item3
> > >
> > > Is it wrong assumption to expect xmlHashScan calling back exactly in
> > > same order as I added these entries?
> >   It is a wrong assumption. You can't expect any specific
> > order back from xmlHashScan() callbacks.
> Ok, thanks. I'll workaround that then. Btw, is it possible recently
> added hash randomization made
> this "more" visible? I mean I was lucky enough it seems that a very
> limited test case (of two entries, well I know),
> worked with older version and now it fails from time to time.

  yes it's very possible that this added randomization just exposed that
to you, you were just lucky until now :-)

Daniel

-- 
Daniel Veillard      | Open Source and Standards, Red Hat
veillard redhat com  | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | virtualization library  http://libvirt.org/