Hi all, Just a quick progress update on the noscreen work.
Broke original large patch down into mostly atomic commits: Original patch was 6574 lines: 45 files changed, 1173 insertions(+), 2281 deletions(-) New series is 21 commits, which (if applied in the right order) should give a mostly working tree at each stage. I say mostly, as there are some fixes later on in SCM which aren't rolled into this section. New patch over 21 commits: 38 files changed, 1150 insertions(+), 2805 deletions (More changes logically rolled in here, e.g. removing function bodies as we deprecate functions - also some changes deemed unnecessary or incorrect are not included - hence the lower file count.) I've now rebased this, and the 40 (ish) commits after it onto cvs-head! (actually based on the noscreen branch in CVS). With a few minor tweaks, the finished product compiles and appears to work fine (as it did against glist_dev). "noscreen" branch is created in head, and some carefully selected commits from my tree are applied. These produced a fully working tree at each checkin: 2006 line diff over the 8 commits: 20 files changed, 126 insertions(+), 1138 deletions(-) --- Made o_complex_mirror() into o_complex_mirror_world(), made o_mirror() into o_mirror_world(), and converted usage. --- Remove unused function o_complex_mirror2() --- Removed o_..._mirror() functions unused since above changes --- Made o_complex_rotate() into o_complex_rotate_world(), made o_rotate_90() into o_rotate_90_world(), and converted usage. --- Modify o_text_place_rotate() to use world coordinates, and o_text_rotate_world internally. --- Removed o_..._rotate() functions unused since above changes. --- Modify o_move_end_lowlevel() to not require screen coordinates as a parameter (which were unused). Alter usage in o_move_end() to match. --- Removed o_..._translate() functions which are unused. --- My git repo: http://www2.eng.cam.ac.uk/~pcjc2/geda/pcjc2.git/ SORRY to anyone who was using the repo - all the SHA1 have changed due to rebaseing. (I did warn I was going to do that - and it will probably happen again!) Heads: R cvs-glist_dev \ R cvs-head | Remote CVS tracking branches R cvs-noscreen / drawing_with_args JUNK 1noscreen_total Initial commit of existing noscreen work when starting with git. The diff is FAR too big for one commit! 2fix_complex_bounds Stupid branch name, but has one commit towards the finished result. 3cache_world_bounds Changes which switch to using cached world bounds instead of continual calculation on demand. Also some fixes which should eventually get propagated down into the commits underneath. noscreen_incremental A branch which splits changes from 1noscreen_total into small atomic bits. noscreen_last_monolith The remaining (fairly big) change where everything switches to calc'ing world coords instead of screen coords. (MIGHT) be possible to split further. noscreen_inc_tests Test application of certain commits from noscreen_incremental. Some re-ordering noscreen_inc_tests2 Test application of certain commits in a useful order to produce a working tree at each stage. noscreen_rebase1 A rebase of the complete work against noscreen_incremental rather than 1noscreen_total precvs-noscreen Selected patches destined for CVS. Used as a working area to produce patches ready for applying via CVS. >precvs-noscreen_mergetest Where the complete branch rebase against cvs_head lives pb_vs_cvs Old attempt at rebaseing PB's patches to be more current against CVS. This is based _before_ the glist merge. -- 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 [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
