--- On Sun, 13/2/11, Shing Hing Man <matmsh at yahoo.com> wrote:
> From: Shing Hing Man <matmsh at yahoo.com>
> Subject: Re: [R] Linked List in R
> To: "Martin Morgan" <mtmorgan at fhcrc.org>
> Date: Sunday, 13 February, 2011, 4:49
> Hi Martin,
> ? Thanks for the pointer ! It seems quite involved to
> create a
> a linked list in the oo way. I need to read up on
> environment.
>
> Thanks!
> Shing
>
>
> --- On Sun, 13/2/11, Martin Morgan <mtmorgan at fhcrc.org>
> wrote:
>
> > From: Martin Morgan <mtmorgan at fhcrc.org>
> > Subject: Re: [R] Linked List in R
> > To: "Shing Hing Man" <matmsh at yahoo.com>
> > Cc: r-help at r-project.org
> > Date: Sunday, 13 February, 2011, 4:10
> > On 02/12/2011 11:08 AM, Shing Hing
> > Man wrote:
> > > Hi,
> > >???I am trying to create a linked list
> > in R by defining a class Node which has a instance
> variable
> > Node.
> > >
> > > setClass("Node",
> representation(rate="numeric",?
> > nextNode="ANY"))
> > >
> > > The above works. But the following gives me a
> warning
> > message.
> > > setClass("Node",
> representation(rate="numeric",?
> > nextNode="Node"))
> > >
> > >
> > >>
> > >> setClass("Node",
> > representation(rate="numeric",? nextNode="ANY"))
> > > [1] "Node"
> > >> removeClass("Node")
> > > [1] TRUE
> > >>
> > >> setClass("Node",
> > representation(rate="numeric",? nextNode="Node"))
> > > [1] "Node"
> > > Warning message:
> > > undefined slot classes in definition of "Node":
> > nextNode(class "Node")
> > >>
> > >
> > > In the case when nextNode is type Node, is it
> possible
> > to get ride of the above? "undefined slot classes"
> > warning message ?
> > >
> > >
> > > Thanks in advance for any assistance!
> >
> > Hi Shing --
> >
> > Not really an answer to your question, but I guess
> linked
> > lists give you
> > constant insert / linear search. The only data type in
> R
> > with constant
> > insert time is an environment. Environments map keys
> to
> > values, with
> > keys being character scalars. So (untested)
> >
> > setClass("Node",representation(next="character",
> > value="ANY"))
> > setClass("LinkedList",
> > ? ? representation(store="environment",
> > head="character")))
> >
> > environments are tricky because of their
> pass-by-reference
> > semantic, and
> > to initialize the LinkedList class you'll want to
> write an
> > initialize
> > method that creates a new environment for each
> instance
> >
> > setMethod(initialize, "LinkedList",
> > ? ? function(.Object, ...,
> > store=new.env(parent=emptyenv()))
> > {
> > ? ? callNextMethod(.Object, ..., store=store)
> > })
> >
> > you'd then write methods on the LinkedList class to
> insert
> > / delete /
> > etc. Likely some new concepts in there but hopefully
> that
> > helps...
> >
> > For the node values, creating S4 objects can be very
> > expensive and if
> > you are thinking of large linked lists that are
> frequently
> > updated, you
> > might do well either to use simple lists or vectors
> to
> > represent the
> > node values, or to rethink whether linked lists are
> what
> > you need --
> > they're not a very friendly R data structure.
> >
> > And since the LinkedList class is based on an
> environment,
> > it has
> > reference semantics that will confuse R users.
> >
> > Martin
> >
> > >
> > > Shing
> > >
> > >
> > >
> > >
> > > ______________________________________________
> > > R-help at r-project.org
> > mailing list
> > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> > > and provide commented, minimal, self-contained,
> > reproducible code.
> >
> >
> > --
> > Computational Biology
> > Fred Hutchinson Cancer Research Center
> > 1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109
> >
> > Location: M1-B861
> > Telephone: 206 667-2793
> >
>
>
>
>
