I was thinking that the RObjectTables appeared before the active
binding stuff was introduced. On quick examination, it seems the
active binding mechanism might be a more stable way to go. That's how
it was done in filehash.
-roger
On Tue, Jul 24, 2012 at 9:15 AM, wrote:
> In my original reply
In my original reply I wrote
The
facility in its current form does complicate the internal code and
limit some experiments we might otherwise do, so I would not be
surprised if it was at least substantially changed in the next year or
two.
This is obviously _not_ the time to invest effort in d
Maybe it is worth considering to document this functionality a bit
more, or expose some wrappers in R? It's a bit obscure right now,
which seems both dangerous in terms of maintenance and a missed
opportunity (especially if people are already building on it).
On Tue, Jul 24, 2012 at 2:06 AM, M
Luke,
Please keep me advised on this, because the Qt interfaces heavily rely on
the ObjectTables (btw, it has worked great for my use cases).
Michael
On Fri, Jul 20, 2012 at 7:32 AM, wrote:
> I believe everyone who has worked on the relevant files has tried to
> maintain this functionality, bu
Hi Jeroen
I am restoring the packages on the Omegahat site as people
ask for them.
I just put the RObjectTables package up in in its current form at
http://www.omegahat.org/RObjectTables/
and am happy to enhance it as needed.
We can chat off-list, if you want.
D.
On 7/19/12 10:30 AM,
I believe everyone who has worked on the relevant files has tried to
maintain this functionality, but as it seems to get used and tested
very little I can't be sure it is functional at this point. The
facility in its current form does complicate the internal code and
limit some experiments we migh
You might be able to take advantage of the ObjectTable support in C, which
is part of R, see R_ext/Callbacks.h.
Michael
On Thu, Jul 19, 2012 at 10:30 AM, Jeroen Ooms wrote:
> I was wondering if anyone knows more about the state of RObjectTables. This
> largely undocumented functionality was intr
I was wondering if anyone knows more about the state of RObjectTables. This
largely undocumented functionality was introduced by Duncan around 2002
somewhere and enables you create an environment where the contents are
dynamically queried by R through a hook function. It is mentioned in R
Internals
