On Wed, 2006-12-13 at 18:24 +0100, Patrick Bernaud wrote:
> I have played with the patches you posted. And while I failed to apply
> them to the glist_dev branch, I think there is not much problems
> between your changes and mines.
I'm sorry about that, they are not completely in sync:
Added new function world_get_single_object_bounds to calculate
the bounds of a single object, and make world_get_complex_bounds
call it.
Revision Changes Path
1.28 +63 -27 eda/geda/gaf/libgeda/src/o_complex_basic.c
That commit may cause you grief. It shouldn't be insurmountable though.
There may be others.
I was going to re-base, but after hearing of your changes, didn't want
to until I knew better what was going to happen about this all. (I still
don't know!)
> The most visible (and probably the biggest) is the changes to the
> get_bounds_* functions to take an OBJECT instead of a LINE, ARC. Apart
> from that it will be no big deal to merge.
See my last email to the dev-list. There are bits of manual merging
thought almost every file we both changed. Not all complicated, but some
of it is.
More importantly, the order of changes is far less logical, and is more
likely to have introduced bugs. The noscreen changes opened up a range
of refactoring possibilities which make more sense being applied after
the noscreen changes.
Removing the DONT_REDRAW flag was one I was thinking of next - which you
have already done. That relates to the mis-use of redraw functions to
re-calc screen coordinates IIRC.
> I think the mistake is to have based your changes on glist_dev and not
> HEAD: correct me if I am wrong but it had nothing that specifically
> require changes from glist_dev and as such should have been part of an
> independent branch.
In hindsight perhaps so, although basing of head would be just as broken
against your changes. I wanted to do the changes in one-sequence, across
all the _glist_ functions too, ensuring that a logical step-by step
process was followed to ensure little or no breakage.
I based of glist_dev because it was scheduled for merge and hadn't been
altered for some time when I started.
> Anyway I will start merging your changes to my refactoring of
> translate and get_bounds function (the action stuff is not concerned)
> and let you know how it really works.
I'm not sure if this is wasted effort or not. The changes need to be
applied incrementally and step by step to be sure they work properly at
each step. It would be interesting to see how compatible they are
though.
> > [...]
> > The noscreen work / experiment I've done, I think has some useful code.
> > Not all of it is purely noscreen related, however it the later (less
> > noscreen related changes) are all separate, and very small commits, and
> > are made possible / sensible / necessary by the noscreen changes.
> >
> > Collectively, they simplify the codebase significantly. It isn't
> > complete yet, but I think represents a cleaner platform to start future
> > refactoring work.
>
> On this particular point please ask Peter TB Brett to forward you my
> answer to his first private message on this matter.
I asked, but he doesn't have the message any more (some inbox problem
after a power failure). If you have any comments for me, please send.
(Or perhaps re-send Peter B the message you would like him to forward).
Regards,
Peter C.
_______________________________________________
geda-dev mailing list
[email protected]
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
