Re: gEDA-dev: Happy Holidays

[Peter Clifton <pcjc2-at-NOSPAMcam_dot_ac_dot_uk>]

On Thu, 2006-12-21 at 00:54 -0500, Ales Hvezda wrote:
> Hi All,
> 
> 	I might be out of touch for a couple of weeks.	I may have
> somewhat unreliable 'net access, but I can't predict that at this point.
> So, Happy Holidays.  See ya all next year,

Have a great holiday!

> 						-Ales
> 
> PS. There are a few patches that I haven't applied yet (I've seen nothing
>     wrong with them at this point).  I will get to them soon or if somebody
>     feels daring, by all means commit them to CVS HEAD.  

Which patches are those?

1. Were any from me? If they are bug-related, let me know which ones,
and I can apply them.

2. Did you mean any of the glist cleanup from PeterB? (If so, I can get
together with him and apply).

3. Did you mean the suggested 3 of Patrick's? (I've wanted to catch you
on IRC to discuss those).

With regard to that sub-set of Patrick's patches - suitably rebased, I
don't think it will be a problem to mould noscreen to fit with those. It
could even make some things easier.

I am getting along nicely with cleaning up my noscreen changes. (Still
about 4000 lines of diff file left to break up into atomic, working
commits.

A few discussion points in the lists recently have shown me that I was
wrong to cull o_complex_get_bounds from my code (which I did due to
duplicated code and functionality with one of the list functions).

Instead it should be a wrapper which calls the appropriate list-bounds
function. It is fairly easy for me to use Patrick's implementation
there, however I wasn't sure how to go about it:

1) Will Patrick mind? - I guess I'll have to ask him... its his code, he
may want to commit it himself.

2) I've already been slated by PB (and caught out) by not working
against CVS HEAD once... I don't want to presume his changes will apply,
then work on top of them - it feels like not learning a lesson from the
previous experience!

On the other hand though, If I use his implementation of a given
function, (good in its own right), to save writing my own - nearly
identical one, merging with any later application of Patricks
refactoring will be much easier.

3) At the moment, I've left a duplicate implementation - which isn't
ideal.


(PS.. I've not re-based on HEAD yet, but the atomic commmits / diffs
I've been producing will be far more likely to work in this respect.)

I don't want this repo "public" yet, (lots of mess), but you can see the
changes at www2.eng.cam.ac.uk/~pcjc2/geda/pcjc2.git/

FYI: Tested to work incrementally are changes in the branch
noscreen_inc_tests.

I'm thinking I may want to branch in CVS, then apply most of these
(rebased against CVS HEAD). (Not the bottom one or two though which
comment out PNG printing code and remove old comments from it).


BRANCHES:

R ptbb-glist-transform    | (Remote branches looking at PeterB's stuff)
R ptbb-master             |
R ptbb-patrick-transform  |
R ptbb-ptbb-selection     |

R cvshead                 | (Remote branches tracking CVS 
R glist_dev               |  - provided by PeterB)

  master                  - My master branch (adds .gitignore files)
                              and cuts down the master makefile
                            Based on glist_dev.

  1noscreen_total         - Result of applying my initial big patch on master

  2fix_complex_bounds     - Only ended up with one minor change, removing
                              some un-necessary recalc's. Was based on 1noscreen_total
                              which has since changed.. 
                              1d69dfed8bd8bde00f5cf8b839cb97148cec2a80 is the start

  3cache_world_bounds     - Changes to use the cached bounds (now in world-coords) rather
                              than computing it each time. Also has some other commits
                              which can be re-ordered to a more appropriate place.
                            Based upon 2fix_complex_bounds.

  drawing_with_args       - An avenue I looked at briefly, passing args to xor drawing funcs,
                              rather than using TOPLEVEL. Perhaps for a later re-factor.
                            Based upon 2fix_complex_bounds.

  noscreen_incremental    - Starting at master, this branch is my current playground... it aims
                              to reach the same codebase as 1noscreen_total in incremental, atomic
                              commits. (Not necessarily in the final application order though).
                            Where I've introduced differences between this and 1noscreen_total which
                              I really do want, I've temporarily committed those on the HEAD of
                              1noscreen_total.
                            I will re-base the other branches on the end of noscreen_incremental when
                              it is ready.

  noscreen_inc_tests      - Cherry-picked commits from noscreen_incremental in a possible application
                              order. Each is tested (superficially) to still compile and work.
                            Based on master. I expect to try against CVS HEAD soon too.

  pb_patch_test           - Testing application of PB's patches with glist code - based on ptbb-master
                              (before PeterB's glist merge patches got into CVS HEAD). Didn't get far.
  pb_vs_head              - Testing application of PB's patches against CVS HEAD - before glist work merged.
  pb_vs_ptbb-master       - Attempt at merging pb_vs_head against changes in pb-master (as above).


-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)



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