Re: gEDA-dev: glist-dev without screen coordinate caching

[Peter TB Brett <peter-at-NOSPAMpeter-b_dot_co_dot_uk>]

On Saturday 16 December 2006 08:38, Patrick Bernaud wrote:

> These functions are working on a OBJECT or a list of OBJECTS that may
> or may not be COMPLEXes are in a file named o_complex_basic.c. For
> these kind of function, we have o_basic.c and o_list.c.
>
> get_object_list_bounds() is a rename of an existing function that
> basically did the same but had complex in his name. On this point this
> is an improvement. But renaming it and leaving it where it is is
> definitely not.
>
> What is a function working on a generic OBJECT is doing in
> o_complex_basic.c?
>
> Even worse, instead of fixing the problem, some other totally
> unrelated functions have been added there too (get_bounds fro glist).
>
>  > [...]
>  > The noscreen branch actually removes all get_bounds functions in screen
>  > coordinates, so the complex won't miss its lack of a function.
>
> We are working on pseudo-objects. Why, for the time between this merge
> and your merge for no-screen, the complex object will not have a
> screen-coords get_bounds function while every other object has? It had
> had one until a few weeks ago.

Hi Patrick,

I'm pretty certain you've got this already in your personal SCM.  If you can 
export ***incremental*** patches from your SCM between some known CVS 
revision and where you want the tree to be, I can then load them up and 
rebase them for you.

My biggest concern is that your development tree is so far diverged from what 
you said your initial point was (tag 'rel_20061020', I believe) that for you 
to merge with head is going to be something of a nightmare, with a high 
probability of causing regressions.  Even doing the manual merge in the 
patchset I just posted was painful, with several tricky conflicts in libgeda 
and gschem.

Fortunately having done so, it was fairly straightforward to rebase the rest 
of glist_dev on top of the result, and I'm just compiling it to test that 
Carlos' transform-while-moving code still works.

Is there any particular reason *why* you don't keep your working repository 
fully merged with CVS HEAD?

Peter

-- 
Fisher Society committee                    http://tinyurl.com/o39w2
CUSBC novices, match and league secretary   http://tinyurl.com/mwrc9
CU Spaceflight                              http://tinyurl.com/ognu2

v3sw6YChw7$ln3pr6$ck3ma8u7+Lw3+2m0l7Ci6e4+8t4Gb8en6g6Pa2Xs5Mr4p4
  hackerkey.com                                  peter-b.co.uk

PGP signature

_______________________________________________
geda-dev mailing list
geda-dev@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev