Re: how to suppress carefully a recursive tree

From: fj <francois.jacq_at_irsn.fr>
Date: Wed, 23 Jan 2008 06:28:54 -0800 (PST)
Message-ID: <beb74362-5853-4b48-b8ea-b87c351b0b7b_at_d4g2000prg.googlegroups.com>


On 23 jan, 12:02, "David Cressey" <cresse..._at_verizon.net> wrote:
> "fj" <francois.j..._at_irsn.fr> wrote in message
>
> news:4e22c235-3e07-4fcd-a9d1-f3362ea99239_at_v46g2000hsv.googlegroups.com...
>
>
>
> > On 23 jan, 01:12, David BL <davi..._at_iinet.net.au> wrote:
> > > On Jan 23, 1:10 am, fj <francois.j..._at_irsn.fr> wrote:
>
> > > > On 22 jan, 15:52, Jan Hidders <hidd..._at_gmail.com> wrote:
>
> > > > > On 22 jan, 12:04, fj <francois.j..._at_irsn.fr> wrote:
>
> > > > > > I know how to suppress a normal tree but I meet the following kind
> of
> > > > > > situation :
>
> > > > > I'm guessing that when you say "suppress" you mean "represent in a
> > > > > database". Correct?
>
> > > > No : I want to destroy, remove, kill ... a part of the data (a
> > > > complete tree or just a branch), but without destroying data shared by
> > > > other trees or branches.
>
> > > Would the mark and sweep algorithm suit you purposes?
>
> > No
>
> > the mark and sweep algorithm needs to know all the tree roots in order
> > to mark all the used objects.
>
> > In my example I indicates that the node b3 belongs to the root r2
> > just for information to explain the storage count. But the deletion
> > routine I want to write just receives "r1" as argument, nothing else.
> > It does not know that r2 exists too ... It is just able to detect the
> > existence of other objects via the storage count of b3 which is a
> > little bit too high for being just referenced by r1 !
>
> How is the tree represented? If it's in the usual child-parent
> relationship, then you will have to traverse the sub tree in order to locate
> all its nodes.

Each parent knows its children.
It could be possible to add extra information in children so they will know theirs parents too.

Yes, at the moment I have to traverse the sub tree in order to locate all its nodes Received on Wed Jan 23 2008 - 15:28:54 CET

Original text of this message