TypedControls branch? was Re: [Clam-devel] Bug solved for deleting Out In Ports and Out In Controls
Pau Arumí
parumi at iua.upf.edu
Thu Jul 17 09:19:56 PDT 2008
On dj, 2008-07-17 at 12:56 -0300, Hernán Ordiales wrote:
> On Thu, Jul 17, 2008 at 12:39 PM, Pau Arumí <parumi at iua.upf.edu> wrote:
> > On dj, 2008-07-17 at 10:49 -0300, Francisco Tufró wrote:
> >> Hi, if we take this as a bug/feature that needs to be implemented, it
> >> also needs to be implemented in typed controls.
> >> I'll add this to the next patch.
> >
> > Hi Francisco,
> > TypedIn|OutControl<T> derives from In|OutControl, right? In that case
> > there is nothing to be done. It is all base-class behavior.
>
> nope, in the current patch implementation there is a
> BaseTypedIn|OutControl and In|OutControl inherits from
> TypedIn|OutControl<TControlData>, anyway i think is just a matter to
> update the code related to BaseTypedIn|OutControl in the same way
> Ferrarn did.
ah, yes. BaseTyped*Control have CanConnect() interface. Yes synching
that patch is not a problem.
> BTW, did you checked the last mails of the thread:
> "TypedControls: asking for advice and opinions"? new problems with
> some internal implementations (needed for a complete migration) arose
> and we were not completely sure about the solutions of some issues...
I was not sure if all was addressed or not. Could your or Francisco send
a brief list of pending issues to be addressed?
> Francisco: please send/resend the patch to the list (although could be
> incomplete at the moment), just to have a better picture of what we
> are talking about...
The TypedControls stuff has shown to be much more complicated and have
many ramifications than initially planed. I fear that managing this big
work with patches is not efficient. On the same time, we are dealing
with base class substitutions and so we can not commit till the whole
port is complete.
Therefore, I see all the ingredients to work with a branch (a
not-keen-about-subversion-branches person is speaking)
>From some weeks ago I'm trying to have a launchpad/bazaar CLAM svn
import, but it is proving difficult:
https://code.launchpad.net/clam
So I propose to create a branch in our subversion. This branch should
live only during the controls transition, and the maintainers should
keep all the changes in trunk commited in the branch. That is, in
permanent state of merge with trunk.
What do you think?
P
More information about the clam-devel
mailing list